NB 5.0GA Pack NB_50_7_M README                                  December 5, 2006
Requirement: NB_CLT_50_7_M
================================================================================
** All 5.0 MP7 Maintenance Packs for NetBackup server, client, all Add-on
products, Database Agents, and Java GUI must be applied so that the entire
system is at the same Maintenance Pack level. **

This Maintenance Pack provides fixes for Veritas NetBackup (tm) UNIX servers.
NetBackup UNIX Add-on products, Database Agents and Java GUI have separate
Maintenance Packs.  

Symantec recommends that the backing up of active file systems be avoided,
or the use of snapshot technologies be implemented. The directory structure
reported back from the file system (to NetBackup) may not contain all of the
files available during the time of backup. NetBackup will not report errors
in many cases where the file's existence is not known to NetBackup as reported
by the file system.


=================
PACK DEPENDENCIES
=================

    -- NB_CLT_50_7_M_<6 digit number>.tar must be installed before this
       Maintenance Pack is installed.

    -- Installation of this Maintenance Pack requires version 1.17.4.40 of
       the Vrts_pack.install script.

    -- Deliverables for the agents or options such as DB2, Oracle, VSM do not
       always change between patches and could result in a patch not being
       delivered.  When upgrading or patching client software, any agent
       software must be upgraded or patched to the latest or matching level at
       the same time as the server and client software.


I.   DOWNLOAD INSTRUCTIONS
II.  KNOWN ISSUES
III. INSTALLATION INSTRUCTIONS
IV.  UNINSTALL INSTRUCTIONS
V.   DESCRIPTION OF PROBLEMS FIXED
      Current Pack
         NB_50_7_M

      Pack History
         NB_50_6_M
         NB_50_5_M
         NB_50_4_M
         NB_50_3_M
         NB_50_2_M
         NB_50_1_M


=========================
I. DOWNLOAD INSTRUCTIONS
=========================
1) Download the NB_CLT_50_7_M_<6 digit number>.tar and
  NB_50_7_M_<6 digit number>.<server>.tar files into the
  /tmp directory,

  where <6 digit number> is an internal tracking identifier

  where <server> is alpha_5, hp_ux, linux, rs6000, sgi, solaris
                   
    NOTE: NB_CLT_50_7_M_<6 digit number>.tar has the client binaries and
    NB_50_7_M_<6 digit number>.<server>.tar has the server binaries and BOTH
    must be installed.

2) Extract the NB_CLT_50_7_M_<6 digit number>.tar and the
  NB_50_7_M_<6 digit number>.<server>.tar files.
       tar xvf NB_CLT_50_7_M_<6 digit number>.tar
       tar xvf NB_50_7_M_<6 digit number>.<server>.tar

       NB_50_7_M will create the files:
       VrtsNB_50_7_M.README
       VrtsNB_50_7_M.<server>.tar.Z
       VrtsNB_50_7_M.postuninstall
       VrtsNB_50_7_M.postinstall
       VrtsNB_50_7_M.preinstall
       Vrts_pack.install

       NB_CLT_50_7_M will create the files:
       VrtsNB_CLT_50_7_M.README
       VrtsNB_CLT_50_7_M.tar.Z
       VrtsNB_CLT_50_7_M.postuninstall
       VrtsNB_CLT_50_7_M.postinstall
       VrtsNB_CLT_50_7_M.preinstall
       Vrts_pack.install



=================
II. KNOWN ISSUES
=================
The following are known issues with NB_50_7_M.  Please note that
these known issues will NOT be resolved after installing this pack.

Description:
   An issue has been identified on certain operating and file system
   combinations when NetBackup does an archive operation. The target files
   are backed up properly, but not all of them are removed from the client.
   The problem is in the underlying operating system and not NetBackup. When
   a directory contains many files, and a large number of consecutive
   unlink()s are issued, the system may incorrectly report that there are no
   more files available. When NetBackup attempts to delete the parent
   directory, the process will fail and may cause the archive operation to
   complete with only partial success.

   This is not a data loss issue. The work-around is to manually delete the
   backed up files from the client after the archive operation completes.

Description:
   Solaris 10 (Sparc) first appeared as a client choice in
   NetBackup 5.0 MP6.  This was in preparation for when Sun Solaris 10
   is generally available and support for Solaris 10 will begin after
   Solaris 10 (Sparc) is officially released.  Solaris 10 support is
   available for the client only.

Description:
   When patching AIX NetBackup Servers, make sure that no
   NetBackup GUIs running on remote systems are monitoring
   the AIX system being patched as this may lead to text busy
   errors.

Description:
   When applying pack updates to AIX host <A>, do not use a
   NetBackup GUI running on any other host to monitor the
   NetBackup jobs/processes/daemons/devices on host <A> during
   the installation.  On AIX this may cause text busy errors
   to occur during the installation.  The safest approach is to not
   have any GUIs running on remote systems which are accessing AIX
   host <A> during pack application.

--------------------------------------------------------------------------------
Description:
   A backup that stops and resumes can successfully complete, but there may be
   multiple duplicate fragments.  This type of problem causes the import to not
   be able to determine which duplicate fragment is correct.  This is
   technically not a data-loss issue (since the data is on tape), however the
   data will be difficult to recover if the import is needed.

--------------------------------------------------------------------------------
Description:
   When using an advanced client snapshot method for backup, the checkpoint
   restart feature should not be used. Doing so may result in a backup error
   due to an attempt to create a new snapshot when a snapshot already exists
   from the original incomplete backup.



===============================
III. INSTALLATION INSTRUCTIONS
===============================
NOTE: Click on the Download Now link, near the bottom of this document
prior to running the following installation procedure for this pack.


For Maintenance Pack installation on a UNIX Cluster Environment:

1) Ensure that prior to installing the maintenance pack, NetBackup is at
  release level 5.0 and configured to run in a cluster.

2) Freeze the NetBackup group (This will avoid a 'failover' during a patch
  installation).

3) Install this Maintenance Pack on the inactive node(s) of the cluster
  (follow the steps below).

4) Install this Maintenance Pack on the active node of the cluster (follow
  the steps below).

5) Unfreeze the NetBackup group.


--------------------------------------------------------------------------------

As root on the NetBackup Master Server:

1) Install NB_50_7_M and NB_CLT_50_7_M Maintenance Pack binaries.

       cd /tmp
       /bin/sh Vrts_pack.install

  NOTE: Selecting the server Maintenance Pack will automatically install the
  client maintenance pack if the client (CLT) .Z file and the README exist in
  the installation directory. The server install will fail if the (CLT)
  .Z file and the README are not present and the CLT pack has not been
  previously installed. The client maintenance pack will NOT be installed
  automatically during a reinstall of the server maintenance pack.  

2) Restart daemons.
       /usr/openv/netbackup/bin/initbprd
       /usr/openv/volmgr/bin/ltid -v

3) The pack install logs can be found in /usr/openv/pack/pack.history once the
  installation is complete.


Additional Notes:

If non-root administrators use the GUI only, the nonroot_admin
script no longer needs to be run.  If the non-root administrators
use the command line or bpadm, the group and file permissions
will have to be changed manually on the NetBackup binaries.
Users can write their own script.  The script is being phased
out because there is a slight security risk that non-root users
may be able to execute NetBckup commands only because those users
are part of a group that is allowed to execute NetBackup commands.

For Solaris Solaris10 to show when configuring clients into a backup
policy, the following command must be executed:

  /usr/openv/netbackup/bin/goodies/new_clients



===========================
IV. UNINSTALL INSTRUCTIONS
===========================
Note:  This will ONLY uninstall the Maintenance Pack from your local machine.

   1) Close the NetBackup user interfaces.

       Make sure the NetBackup server has no active jobs running (for
       example, backups, restores, or duplications).

       If a database agent is being used, such as Oracle,
       ensure that the database services are stopped.

   2) Change directory to the patch save directory.  
       Substitute the pack name for ${PACK} in the following command:

          cd /usr/openv/pack/${PACK}/save

   3) Run the un-install script:

         ./Vrts_pack.uninstall

   4) Verify that the pack uninstalled successfully by checking
       /usr/openv/pack/pack.history.

   5) If update_clients was run after the pack was originally INSTALLED,
       run it again after that pack is successfully UNINSTALLED.

   6) If necessary, restart the NetBackup and Media Manager daemons:
       /usr/openv/netbackup/bin/goodies/netbackup start

   

=================================
V. DESCRIPTION OF PROBLEMS FIXED
=================================
The following are descriptions of the problems fixed.  Please read the entire
document before installing.

README Conventions:

Description
    Describes a particular problem contained in this pack.

** Description **  
    Describes a problem that can lead to potential data loss. Please
    read these problem descriptions carefully.

Workaround
    Any available workarounds to a problem are also listed. Workarounds
    can be used INSTEAD of applying the patch, however, Symantec strongly
    recommends that you are at the latest patch level.

Additional Notes
    Any additional information regarding a problem is included.


=============
Current pack
=============

================================================================================
Etrack Incident = ET512621

Description:
   Bptm crashed when multiplexed backups were duplicated when the
   Preserve multiplexing function was set.

Workaround:
   To avoid this issue, perform one of the following:
   
   - Create the following file on all media servers:
      UNIX: /usr/openv/netbackup/db/config/DISABLE_READ_FRAG_CHK
      Windows: <install_path>\NetBackup\db\config\DISABLE_READ_FRAG_CHK
   
   OR:
   
   - Duplicate the backups without setting the Preserve multiplexing function.
--------------------------------------------------------------------------------
Etrack Incident = ET500462

Description:
   Fixed an error that caused bpsched to loop in the find_next_stunit
   function after an Inline Tape Copy (ITC) multi-copy failure occurred.
--------------------------------------------------------------------------------
Etrack Incident = ET510937

Description:
   An issue with the HP-UX compiler caused it to optimized out references
   to the shared memory control variables.  This issue caused restores to
   fail with the following errors in the job log and tar debug log:
   
   job log:
   22:16:27 (4753.046) data buffers out of sequence, expected number 157884,
   received 157886
   23:32:46 (4753.053) data buffers out of sequence, expected number 404927,
   received 404932
   
   debug log:
   22:16:27 (4753.046) data buffers out of sequence, expected number 157884,
   received 157886
   23:32:47 (4753.053) data buffers out of sequence, expected number 404927,
   received 404932
--------------------------------------------------------------------------------

Etrack Incident = ET417916

Description:
   A backup would run multiple times, ignoring the number of tries for
   FlashSnap streams.  
--------------------------------------------------------------------------------
Etrack Incident = ET516641

Description:
   A bpverify of a successfully imported image no longer returns a 191 error.
--------------------------------------------------------------------------------
Etrack Incident = ET519760

Description:
   The partial restore of a database backup failed.  For example, when
   restoring a single, Oracle tablespace from a backup set, the following
   types of messages were written to the bptm log:
   
   14:50:59.752 [19995] <2> read_brm_msg: STOP RESTORE algol_1134518480 EXIT=0
   14:50:59.752 [19995] <2> process_brm_msg: GCACTIVE but CHILD_PID not alive
   
   In addition, the following messages were written in the dbclient log:
   
   14:50:59.718 [1932.1552] <4> closeApi: INF - FORCED EXIT STATUS 0:
     the requested operation was successfully completed
   14:50:59.718 [1932.1552] <4> closeApi: INF - closing commSock 220
   14:50:59.718 [1932.1552] <4> closeApi: INF - Do not close dataSock
   14:50:59.718 [1932.1552] <4> closeApi: INF - closing nameSock 952
   14:50:59.718 [1932.1552] <4> serverExitStatus: entering serverExitStatus.
   14:50:59.718 [1932.1552] <4> readCommMessages: Entering readCommMessages
   15:06:04.724 [1932.1552] <16> readCommFile: ERR - timed out after
   900 seconds while reading from C:\Program Files\VERITAS\NetBackup\Logs\
     user_ops\dbext\logs\1552.0.1135025111  
--------------------------------------------------------------------------------
Etrack Incident = ET409359

Description:
   Multi-streaming with wild-cards and NEW_STREAM directives did not process
   correctly.
--------------------------------------------------------------------------------
Etrack Incident = ET426095

Description:
   A single-stream, non-checkpoint, Windows Open File Backup (WOFB) with a
   134 status would lose the snap information.
--------------------------------------------------------------------------------
Etrack Incident = ET515430

Description:
   The bpbackup command would receive a status code 0 when the backup was
   canceled.  (This applied to Linux systems only.)  
--------------------------------------------------------------------------------
Etrack Incident = ET507223

Description:
   Veritas Volume Snapshot Provider (VSP) files are no longer left after a
   134-status re-queued backup was cancelled or the window was closed.
--------------------------------------------------------------------------------
Etrack Incident = ET497000

Description:
   A backup was being initiated incorrectly when a STREAMS file could not be
   opened.
--------------------------------------------------------------------------------
Etrack Incident = ET515410

Description:
   During the creation of the binary catalog, a flaw in the way that the
   user name and group name lookup table was accessed caused the file to
   grow excessively large.

Additional Notes:
   The problem is most likely to occur on Linux.
--------------------------------------------------------------------------------
Etrack Incident = ET519237

Description:
   The commands, (bpexpdate -recalculate and bpexpdate -backupid), allow
   image expiration dates to be changed while an image duplication is in
   progress.  If the image spans tapes, this can result in different
   expiration dates for the individual fragments.
   
   The etrack is to patch bpdbm to fail an image expiration change when a
   duplication is in progress.  

Workaround:
   To avoid this issue, do not change the expiration dates on images while
   image duplication is in progress.  
--------------------------------------------------------------------------------
Etrack Incident = ET431394

Description:
   Non-multiplexed duplication of a True Image Recovery (TIR) backup written
   to tape would fail if the last fragment was zero bytes in length.
--------------------------------------------------------------------------------
Etrack Incident = ET522062

Description:
   Restore to an alternate client (not a master or media server) by a SAN
   media server failed with an error status 159:
   
   The licensed features for this system did not permit the backup of a
   remote client (<client>).  To avoid this issue, make sure that the
   license key is entered properly.
   
   SAN media server licensing should be enforced for backup but not for
   restore to allow maximum flexibility for device recovery and data
   restoration.  
--------------------------------------------------------------------------------
Etrack Incident = ET432238

Description:
   Corrected ACS timeouts because ACSLS server had communication problems
   with media server.  To correct this issue, a 20-second wait was removed
   before checking for drive readiness on an ACS mount.
--------------------------------------------------------------------------------
Etrack Incident = ET520521

Description:
   A stack-based buffer overflow vulnerability existed in the volume manager
   daemon (vmd) running on NetBackup servers. If an attacker was able to
   gain access to a vulnerable NetBackup server and successfully exploit this
   issue, it could have lead to arbitrary code execution and resulted in
   unauthorized access with elevated privileges on the targeted system.
   
   This vulnerability impacted only NetBackup server systems and did not
   impact NetBackup client systems.
--------------------------------------------------------------------------------
Etrack Incident = ET432235

Description:
   Changes were made to correct the use of access bits with the mailslot
   and with the storage slots in robotic libraries.
   
   In addition, added a wait-before-eject for Vault-style ejects for
   libraries that take extra long for the mailslot to become ready after
   being accessed.

Additional Notes:
   Access bit checking may require a new device_mappings file depending on
   the robot type.
--------------------------------------------------------------------------------
Etrack Incident = ET512960

Description:
   During the process of Intelligent Disaster Recovery (IDR), more
   specifically during the installation of the base Windows operating
   system, the Windows server crashed with a blue screen in certain
   conditions.
   
   The blue screen contained the following information:
   
   STOP: 0x0000007B INACCESSIBLE_BOOT_DEVICE.
   
   The root cause of the problem was that the Windows operating system CD
   used to create the bootable IDR media had Service Pack 4 integrated with
   it.  The SIF file that the IDR process used to create the bootable IDR
   media did not.
   
   This problem occurred while performing IDR on a Traditional Chinese
   version of a Windows 2000 Server.  The fix introduced in this pack only
   applies to this combination.
--------------------------------------------------------------------------------
Etrack Incident = ET508894

Description:
   A backup that failed to progress far enough to save its .parms file could
   not be restarted.
--------------------------------------------------------------------------------
Etrack Incident = ET514191

Description:
   The bpdbjobs -resume command reported an incorrect number of restores
   that were resumed.  
--------------------------------------------------------------------------------
Etrack Incident = ET498026

Description:
   Unable to restore files or folders when the user name or group name was
   longer than 32 characters.

Workaround:
   To avoid this issue, do not use user names or group names that are greater
   than 32 characters.
--------------------------------------------------------------------------------
Etrack Incident = ET500514

Description:
   There are end case situations, especially when a master server cluster
   node goes inactive, where the bprdreq -terminate and bpdbm -terminate
   commands in bp.kill_all might take several minutes to complete.  This
   caused inconsistencies in a cluster environment.
--------------------------------------------------------------------------------
Etrack Incident = ET499201

Description:
   A problem existed that caused bpdbjobs -all_columns to take a segmentation
   fault in either the FreeJobs() or the PrintJobs() function with a stack
   overflow.  
--------------------------------------------------------------------------------
Etrack Incident = ET512995

Description:
   Catalog archiving failed if there were backup policies that referred to
   the master server as a client with both fully-qualified and short names.

Workaround:
   To avoid this issue, do not mix short and long names on policies that
   back up the master server.

Additional Notes:
   The special catarc policy that is defined for catalog archiving must
   specify the master server as the policy client.  If the master server
   is configured with a fully-qualified name (such as, myhost.mycompany.com),
   the client name in the catarc policy must also use that fully-qualified
   name.  If the master server is configured with a short name
   (such as, myhost), the catarc policy must also use the short name.
   
   Other backup policies that specify the master server as a client may use
   the short name, even though the master server is configured with a
   fully-qualified name.  
--------------------------------------------------------------------------------
Etrack Incident = ET410059

Description:
   Even when the window in the policy was already open, a user could
   encounter intermittent 199 errors, operation is not allowed.
   
--------------------------------------------------------------------------------
Etrack Incident = ET411649

Description:
   After an Inline Tape Copy (ITC) completed, the Activity Monitor would
   report an incorrect value for the Number of files field in a second copy.
--------------------------------------------------------------------------------
Etrack Incident = ET523673

Description:
   In the latest MP release, DSSU relocation in an SSO environment sometimes
   fails with Status 191.  
--------------------------------------------------------------------------------
Etrack Incident = ET498169

Description:
   Attempting to get the list of running processes from within the Java user
   interface failed on HP-UX 11.11.
--------------------------------------------------------------------------------
Etrack Incident = ET512411

Description:
   A problem existed as a result of bpjobd being too slow to acknowledge a
   bptm connection. (A 60-second timeout was reached and caused bptm to exit
   with 54 and 25 errors.)
--------------------------------------------------------------------------------
Etrack Incident = ET430474

Description:
   With the pack it is possible to import Backup Exec images that have files
   with filenames that contain embedded, new lines in them.
--------------------------------------------------------------------------------
Etrack Incident = ET500471

Description:
   FlashBackup restores failed if there were hard-linked files that had a
   large amount of meta data. The restore failed with a log entry in the
   bprd log similar to the following:
       
   3:07:22.987 [16530] <32> bprd.sfr: sort_blockmap: internal error - currec
   189441 > blk_num 0  

Workaround:
   To avoid this issue, restore one instance of the file and manually
   recreate the hard links. This requires you to remove the hard link files
   from the restore list.
--------------------------------------------------------------------------------
Etrack Incident = ET427219

Description:
   Occasionally, a 134 status code (unable to process request because the
   server resources are busy) would incorrectly be reported as a 24 status
   code(socket write failed).  When this occurred it caused the job to fail
   instead of being retried when resources were available.
--------------------------------------------------------------------------------
Etrack Incident = ET416698

Description:
   The restore of an offhost VxFS_Checkpoint Instant Recovery database backup
   no longer fails.  
--------------------------------------------------------------------------------
Etrack Incident = ET532912

Description:
   Symantec is preemptively instituting changes in all the product with the
   intention of hardening NetBackup against potential security problems.  
--------------------------------------------------------------------------------
Etrack Incident = ET526285

Description:
   In a very large stream-count environment, a work-list creation takes too
   long and causes delays with the jobs going active.
--------------------------------------------------------------------------------
Etrack Incident = ET532656

Description:
   The SG driver caused a kernel panic because of a NULL pointer de-reference.
   The SG driver now logs SCSI CDBs, up to sixteen bytes, instead of
   truncating them at twelve bytes.
--------------------------------------------------------------------------------
Etrack Incident = ET506427

Description:
   Unable to create a schedule in a policy using the New button on a
   Linux SUSE 8.0 or 9.0 system.  To resolve this issue, a newer Sig version
   (Sig2.60.017) was included in this pack.  
--------------------------------------------------------------------------------
Etrack Incident = ET532664

Description:
   A TIR backup will no longer create a corrupted tape image if the amount
   of TIR data received does not match the amount expected.
--------------------------------------------------------------------------------
Etrack Incident = ET533191

Description:
   Running asynchronous I/O on large files lead to delays in servicing
   requests.  NetBackup was designed to handle this, however, an error
   existed such that if a read error was encountered with an errno of
   EAGAIN, NetBackup would treat it as a generic error instead of retrying.
   This uncommon problem caused the backup to fail.  

Workaround:
   To avoid this issue, turn off asynchronous I/O.
--------------------------------------------------------------------------------
Etrack Incident = ET527390

Description:
   There was a potential race condition during log roll-over that caused
   log messages to be erroneously written to a socket rather than the newly
   opened log file.  
--------------------------------------------------------------------------------
Etrack Incident = ET526960

Description:
   A problem occurred when the Archive was run using NBWin.  There was a
   change in the behavior that caused data loss when only one file existed
   within the directory at the start of an archive job and new files were
   added to that directory before the archive completed.  BPCD showed that
   at the end of the job, the archive was called with an rm -rf command that
   removed the directory and all files, resulting in a data loss of the new
   files that were added.
--------------------------------------------------------------------------------
Etrack Incident = ET416950

Description:
   Bpadm would potentially crash when selecting option (t) (Storage Unit
   Group Management), then option (m) (Modify), and then selecting any of the
   modify options.
--------------------------------------------------------------------------------
Etrack Incident = ET427224

Description:
   When performing a multiplexed duplication, occasionally the source tape
   would incorrectly get unmounted.  
--------------------------------------------------------------------------------
Etrack Incident = ET523586

Description:
   Once a day, at midnight, the bprd service on the master server connects to
   the hosts in its SERVER list and starts a process on those servers to
   remove old debug log files.  If a host is in the MEDIA_SERVER list but not
   in the SERVER list on the master server, bprd will not contact that host.
   A change was made to ensure that bprd connects to hosts in both the SERVER
   list and the MEDIA_SERVER list on the master server to remove old debug
   log files.
--------------------------------------------------------------------------------
Etrack Incident = ET431372

Description:
   A job scheduled for every two days no longer runs every 10 minutes.  
--------------------------------------------------------------------------------
Etrack Incident = ET517649 ET526746 ET526747

Description:
   This pack contains a software utility (nbsupport 2.1.1) that creates a
   support package that consists of multiple reports that are most often
   asked for by NetBackup support personnel.  
--------------------------------------------------------------------------------
Etrack Incident = ET534513

Description:
   This pack contains a change to ensure that the parent job of a Disk
   Staging Storage Unit (DSSU) staging process does not hang.
--------------------------------------------------------------------------------
Etrack Incident = ET493283

Description:
   It was possible that a recovered NBAC would not work because bp.conf file
   was not restored properly, thus making NBAC virtually unusable.

Workaround:
   To avoid this issue, manually restore bp.conf.
--------------------------------------------------------------------------------
Etrack Incident = ET518249

Description:
   The credentials for a user that belongs to a nested global group were
   missing group details.
--------------------------------------------------------------------------------
Etrack Incident = ET530317

Description:
   The bpcoverage binary in NetBackup 5.0 MP4 would cause a crash.  

Workaround:
   To avoid this issue, use a pre-NetBackup 5.0 MP4 bpcoverage binary.
--------------------------------------------------------------------------------
Etrack Incident = ET524765

Description:
   When bprestore is run with the option to wait for the completion of the
   restore job, a firewall between bprestore and bprd can shut down the
   socket. This causes bprestore to wait for the completion status which
   never comes.

Workaround:
   To avoid this issue, increase the timeout on the firewall.
--------------------------------------------------------------------------------
Etrack Incident = ET538861

Description:
   bpbkar no longer times out while waiting for the shared memory file to be
   created in shared memory backups.
--------------------------------------------------------------------------------
Etrack Incident = ET537454

Description:
   A restore of a large backup image would potentially fail if a large number
   of data buffers (for example, 1024) was configured.  The number of data
   buffers is configured on UNIX systems in
   /usr/openv/netbackup/db/config/NUMBER_DATA_BUFFERS, and for Windows
   systems in <install_path>\NetBackup\db\config\NUMBER_DATA_BUFFERS.

Workaround:
   One of the following may be done to avoid this problem:
     - Set the number of data buffers to a smaller value (For example, 128).
     - Create the following file:
         UNIX: /usr/openv/netbackup/db/config/MIN_RESTORE_SKIP
         Windows: <install_path>\NetBackup\db\config\MIN_RESTORE_SKIP
     - Then enter a number greater than the following:
         (size_data_buffers) * (number_data_buffers) / 512
--------------------------------------------------------------------------------
Etrack Incident = ET536904

Description:
   Restores of millions of files from a FlashBackup were taking a very long
   time because of the way that possible hard-link files were tracked.
--------------------------------------------------------------------------------
Etrack Incident = ET530796

Description:
   BPTM would sometimes corrupt the on-disk image when recovering from a
   disk-full condition.  A backup policy using inline copy and multiplexing
   to a DSU/DSSU was a requirement for customers to encounter this problem.
--------------------------------------------------------------------------------
Etrack Incident = ET538175 ET537660 ET538160 ET537509 ET537473 ET537518 ET537526
ET537535 ET537555 ET537412 ET542502

**Description:
   Multiple buffer overflow vulnerabilities have been identified in daemons
   that run on Veritas NetBackup master, media, and client servers.   An
   attacker, if able to access a vulnerable Veritas NetBackup server and
   successfully exploit these issues, could potentially execute arbitrary
   code resulting in possible unauthorized and elevated privilege access to
   the targeted system.
   
   For more information about this vulnerability, refer to TechNote 281521 on
   the Symantec Support Web site (http://support.veritas.com/docs/281521).
--------------------------------------------------------------------------------
Etrack Incident = ET541873

Description:
   Added support for the new Quantum DLT-S drives.
--------------------------------------------------------------------------------
Etrack Incident = ET541953

Description:
   Robotic Inventory using the vmadm command did not work for ACS robot types.
   The error result that was printed contained the following message.
   
   invalid command usage (4) and unrecognized option -acs_T10000t1
--------------------------------------------------------------------------------
Etrack Incident = ET546691

Description:
   After executing a Vault Duplicate backup, the following warning
   message was displayed in the bpdm and bptm logs:
   
   socket option SO_REUSEADDR failed
--------------------------------------------------------------------------------
Etrack Incident = ET542647

Description:
   Added a -move command into the bpdbm -consistency check to explicitly
   move out the corrupted catalog images. bpdbm -consistency no longer
   moves corrupted images without the option being explicitly specified.
--------------------------------------------------------------------------------
Etrack Incident = ET570698

Description:
   The SRCMEDIAID and DSTMEDIAID fields of jobs database were not being
   written to.
--------------------------------------------------------------------------------
Etrack Incident = ET569902

Description:
   An issue existed that caused ltid to die and report an error if it
   received a message that was greater than 208 bytes.  
--------------------------------------------------------------------------------
Etrack Incident = ET575042

Description:
   Running bpexpdate to change the expiration of a phase I, imported image
   caused it to be deleted if the backup spanned tapes and if the first tape
   had not been imported yet.
--------------------------------------------------------------------------------
Etrack Incident = ET580781

Description:
   Corrected the use of the media mapping ACS_LTO_400G to hcart2 entry in
   vm.conf.
--------------------------------------------------------------------------------
Etrack Incident = ET573315

Description:
   True Image Restore (TIR) backups (including NDMP backups) did not report
   an accurate cause of the failure to write the TIR fragment.
--------------------------------------------------------------------------------
Etrack Incident = ET575707

Description:
   Corrected an issue with the robot inventory update when the only slot left
   in the robot is the last slot.  
--------------------------------------------------------------------------------
Etrack Incident = ET566068

Description:
   The NetBackup restore progress logs cannot grow more than 2GB.  If they do
   it can cause problems for restores with a large number of files.
--------------------------------------------------------------------------------
Etrack Incident = ET570194

Description:
   A multiplexed duplication job failed with the following error.
   
      Amount of data read (0) does not match the fragment kbytes
--------------------------------------------------------------------------------
Etrack Incident = ET581150

Description:
   An e-mail problem on a client caused bpbrm to hang indefinitely in a
   waiting state.
--------------------------------------------------------------------------------
Etrack Incident = ET585294

Description:
   The bpdbm -converti2 command failed to move compressed catalog images in
   catstore when it converted version 1(IDSTRUCT=1) to version
   2(IDSTRUCT=2) in the catalog image directories.
--------------------------------------------------------------------------------
Etrack Incident = ET574023

Description:
   In certain instances under NetBackup Access Control, bpjobd and vmd would
   leak memory.
--------------------------------------------------------------------------------
Etrack Incident = ET593291

Description:
   A fix was added to this pack that corrects the loss of Tape alert
   flags CLEAN_PERIODIC and CLEAN_NOW on certain drives.
--------------------------------------------------------------------------------
Etrack Incident = ET594036

Description:
   A valid image could be overwritten if a job was canceled during a fragment
   break on an AIX, HP_UX, or SGI system.
--------------------------------------------------------------------------------
Etrack Incident = ET578702

Description:
   The UNIX pack installer -i option correctly reports on uninstalled packs.
--------------------------------------------------------------------------------
Etrack Incident = ET594873 ET793503 ET793460

Description:
   Using the catalog archive process, files were not being removed from the
   catstore directory.
--------------------------------------------------------------------------------
Etrack Incident = ET618475

Description:
   A problem existed that caused V_strlcpy to fail to add a terminating
   backslash zero (\0) when the source string is zero length.
--------------------------------------------------------------------------------
Etrack Incident = ET638899

Description:
   NDMP backup would fail with a core dump. A call was fixed so that null
   value would be handled appropriately.
--------------------------------------------------------------------------------
Etrack Incident = ET647546

Description:
   An attempt to restore a database would fail if the database name contained
   Internationalization (I18N) characters.
--------------------------------------------------------------------------------
Etrack Incident = ET703422

Description:
   When the bpexpdate command was run and the recalculate option was used, a
   fault would occur in the bpdbm process.
--------------------------------------------------------------------------------
Etrack Incident = ET767142

Description:
   Received an Allocation Failed message when trying to connect to the
   Activity Monitor.  This occurred when using the Java Activity Monitor.
--------------------------------------------------------------------------------
Etrack Incident = ET700839

Description:
   A change was made to correct excessive resource usage during a synthetic
   backup.
--------------------------------------------------------------------------------
Etrack Incident = ET777641

Description:
   Failing synthetic backups would unintentionally perform a catalog cleanup
   instead of deleting the single failed backup image.
--------------------------------------------------------------------------------
Etrack Incident = ET594980

Description:
   The vmdareq output was truncating the hostname and drivename to three
   characters.
--------------------------------------------------------------------------------
Etrack Incident = ET764807

Description:
   On Linux clients, raw partition backups of raw devices would fail with a
   status 6 error when compression was enabled.

Workaround:
   To avoid this issue, use the block device rather than the raw device.
--------------------------------------------------------------------------------
Etrack Incident = ET649745

Description:
   Warning messages were ignored during user-directed archives which could
   result in the loss of data.  Warnings should have caused the deletion of
   files to be aborted.
--------------------------------------------------------------------------------
Etrack Incident = ET584157

Description:
   Restores of incremental FlashBackups with a Windows master and a Solaris
   client did not restore correctly.
--------------------------------------------------------------------------------
Etrack Incident = ET789208

Description:
   A status 50 error would occur on the Activity Monitor for large
   duplications that take longer than 20 minutes.  However, some duplication
   jobs seemingly failed with a status 50 (shown in the Activity Monitor)
   when in fact the jobs continued to run to success (all images duplicated
   successfully).
   
   Status 50 errors occur only when the first image to be duplicated in the
   batch takes longer than 20 minutes. Smaller images pose no problem, their
   duplication batch job runs through to status 0.

Workaround:
   Technically the duplication still succeeds after the status 50 if you let
   it, it just looks like it fails.
--------------------------------------------------------------------------------
Etrack Incident = ET575347 ET609338 ET609487 ET609540 ET608376 ET609623 ET614668
ET612687 ET616976 ET617837 ET616940 ET622225 ET624432 ET633297 ET641183 ET699951
ET701953 ET781198 ET597210 ET598796 ET575338 ET596635 ET612761 ET787052 ET788824
ET786896 ET814652 ET575351 ET575352 ET575097 ET575330 ET575331 ET575345 ET575337
ET575335 ET577778 ET575333 ET596395 ET575334 ET575313 ET583770 ET575353 ET581028
ET575346 ET575329 ET575343 ET575308 ET575732 ET575309 ET575341 ET577306 ET575350
ET596637 ET596130 ET597780 ET601849 ET596188 ET596645 ET596654 ET596697 ET597281
ET597219 ET597298 ET596342 ET597379 ET598052 ET598865 ET599823 ET598813 ET600384
ET601396 ET601454 ET601344 ET600735 ET602205 ET600844 ET602349 ET603659 ET605285
ET602850 ET575311 ET607666 ET608687  

Description:
   Buffer overflow areas have been identified and corrected in daemons running
   on Veritas NetBackup master, media, and client servers.  Successful access
   to a vulnerable Veritas NetBackup server and the ability to successfully
   execute arbitrary code could potentially result in unauthorized access with
   elevated privilege on a system.
--------------------------------------------------------------------------------
Etrack Incident = ET792319 ET792320 ET843464 ET847074

Description:
   A logic error has been identified and corrected in daemons that run on
   Veritas NetBackup master, media, and client servers.  Successful access to
   and manipulation of this logic error could potentially lead to arbitrary
   command execution with elevated privileges on a vulnerable system.
   
   For customers who need to back out this patch and want to automate the
   down-rev process on their clients using the Update_clients script can
   contact Support to obtain a revised bpinst file for their respective
   version.
   
   The following scenarios do NOT require a new bpinst from Support:
   
   - If, in the future, you are downgrading to 50MP7 or higher, 51MP6 or higher,
       or 6.0MP4 or higher.
   
   - If you are running Windows on your server.
   
   - If you are only downgrading any of the following addons:
     - DB agents (DB2, Lotus Notes, Oracle, SAP, Sybase, or Informix)
     - Encryption (all key lengths)
     - Java
     - Advanced Client
     - BMR master or BMR boot server
     - NDMP
     - Vault
     - GDM (5.x only)
     - Advanced Reporter (5.x only)
     - NetBackup Operations Manager (6.0 only)
     - Storage Migrator
     - Japanese or Chinese Language Packs
================================================================================



============
Pack History
============

=========
NB_50_6_M
=========
Etrack Incident = ET409355 ET263530

Description:
   For Solaris10, a change needed to be made to make sure that /etc/services
   was updated before inetd.conf.  This was only an issue on UNIX servers and
   only if YP was not configured.
--------------------------------------------------------------------------------
Etrack Incident = ET341251

Description:
   If a machine has NetBackup and a Maintenance Pack (MP) installed, and an
   uninstall is performed of the MP that requires a reboot, and NetBackup is
   uninstalled before the reboot, the DLL named msvcp70.dll is removed.  The
   installer has been modified to ignore MFC DLL in the Windows system32
   folder.

Workaround:
   To avoid this issue, perform a reboot when you are requested after
   you uninstall the MP and before you uninstall NetBackup.  
--------------------------------------------------------------------------------
Etrack Incident = ET312138 ET312142

Description:
   NetBackup 5.0 clusters need to be aware of NetBackup Operations Manager.
   Without this patch NetBackup Operations Manager will not work in a
   NetBackup 5.0 cluster.

Workaround:
   The user could add the following lines to
   /usr/openv/netbackup/cluster<cluster_platform>/<platform>_NBU_RSP.
       
   DIR=var mkdir
   DIR=var/global mv
   LINK=var/global  
--------------------------------------------------------------------------------
Etrack Incident = ET352773

Description:
   DSSU relocation would fail after upgrading to NetBackup 50_MP5 or 51_MP3.
--------------------------------------------------------------------------------
Etrack Incident = ET375868

Description:
   A user currently cannot enter more than 99 dates into a calendar
   schedule.  To resolve this issue, increase the size of some of the arrays.  
--------------------------------------------------------------------------------
Etrack Incident = ET369903

Description:
   Disk Staging cleanup was not occurring, nor was it notifying the user
   quickly enough if the Disk Staging Unit was a NAS mounted file system.
--------------------------------------------------------------------------------
Etrack Incident = ET346374 ET322960

Description:
   The display of new ACS vendor media types T10000T1, T10000TS, and
   T10000CT has been corrected.  In addition, corrected the allowable types
   of an ACS vendor media type LTO_400G.  
--------------------------------------------------------------------------------
Etrack Incident = ET361922

Description:
   Some character sets do not support upper case even though the underlying
   language does.  This caused some wildcard constructs to fail because the
   upper and lower cases were not distinguishable.

Workaround:
   To resolve this issue, the wildcard constructs were reworked so [] and '-'
   could not be used with upper and lower case.
--------------------------------------------------------------------------------
Etrack Incident = ET325077

Description:
   Backups of NetLink 2.x and higher files failed.
--------------------------------------------------------------------------------
Etrack Incident = ET355613

Description:
   Backups of an NFS-mounted file system would hang if the NFS server was down.
--------------------------------------------------------------------------------
Etrack Incident = ET380236

Description:
   If the NOexpire touch file was set, then the NetBackup database image
   Cleanup was disabled.  However, media would still expire.  The media no
   longer expires if the NOexpire touch file is set.  Bpexpdate
   -deassignempty, which deassigns media if there are no corresponding images
   in the image database, is disabled.
   
   The NOexpire touch file is set in the following locations:
   
   For UNIX:
   /usr/openv/netbackup/bin/NOexpire
   
   For Windows:
   VERITAS\NetBackup\Bin\NOexpire  
--------------------------------------------------------------------------------
Etrack Incident = ET330684

Description:
   Canceling a queued Vault job caused all backups to cancel.
--------------------------------------------------------------------------------
Etrack Incident = ET321735

Description:
   The use of a closed socket error message was filling the scheduler log.
--------------------------------------------------------------------------------
Etrack Incident = ET367821

Description:
   A user backup could not be cancelled using the activity monitor.
--------------------------------------------------------------------------------
Etrack Incident = ET347553

Description:
   On Windows platforms only: The scheduler would enter an infinite loop after
   A read message queue failure occurred.
--------------------------------------------------------------------------------
Etrack Incident = ET349782

Description:
   The scheduler would enter an infinite loop if all of the following
   occurred:
   - The backup is checkpoint enabled.
   - The backup is twinning.
   - One of the twined copies failed.
   - The non-failed copy exits with a failure status.
--------------------------------------------------------------------------------
Etrack Incident = ET363304

Description:
   Adding a job to a multiplex group failed to check if a HOLD_134 exists on
   the multiplex storage unit. This resulted in excessive 134 status backups.
--------------------------------------------------------------------------------
Etrack Incident = ET366177

Description:
   Field 11, of the 'bpdbjobs -all' output, did not reflect the correct end
   Time of the last try.
--------------------------------------------------------------------------------
Etrack Incident = ET316645

Description:
   The scheduler performance suffered if the media server connection is
   intermittent.  The problem is aggravated when schedule continuously polls
   the media server for NetBackup version information.  
--------------------------------------------------------------------------------
Etrack Incident = ET314631

Description:
   An issue with the scheduler dumping core has been resolved.
--------------------------------------------------------------------------------
Etrack Incident = ET325882

Description:
   The Activity Monitor was continuously displaying the 'positioning' for
   multi-fragment backups.  
--------------------------------------------------------------------------------
Etrack Incident = ET300754

Description:
   The interval-mode backup windows would open longer than six days
   (144-hrs) and resulted in continuous backups on the seventh day, until the
   window closes.  
--------------------------------------------------------------------------------
Etrack Incident = ET301500

Description:
   A Scheduler race condition was occurring: Child sched was sending a
   ‘done’ to the main sched while the main sched was sending a ‘term one’ to
   the child sched. The child sched erroneously treated the ‘term one’ as the
   response to the ‘done’ and then exited prematurely.
   
   The correct child response is to discard (ignore) the ‘term one’
   request and wait for the main sched response to the 'done' command.
   
   Also missing was code to discard (ignore) a ‘suspend’ request after having
   sent a 'done' to the main sched.  
--------------------------------------------------------------------------------
Etrack Incident = ET320556

Description:
   When a resume backup exits, it would attempt to send a request queue
   message to the GUI. But the GUI does not read these messages, so the
   message would stay queued until the main sched exited.
--------------------------------------------------------------------------------
Etrack Incident = ET323182

Description:
   If the stream in control snap failed on a WOFB, then the other streams were
   allowed to continue a non-snap backup. This was the correct operation. The
   problem was that the non-control streams performed a backup of all files in
   the file list, but they should have only backed up their specific files.  
--------------------------------------------------------------------------------
Etrack Incident = ET316634

Description:
   User backups failed to check if an exiting backup was the last backup job
   from a specific SPID before removing all user backups from the work list.
   This oversight caused the removal of active jobs from the work list, and
   resulted in a sched core.
--------------------------------------------------------------------------------
Etrack Incident = ET365012

Description:
   When a C_GET_HOSTINFO request is made from a pre-6.0 installation, the
   [expected] request type and client name are appended to the MAGIC string.
   For 6.0 and later these two fields are sent separately.  Infer the level
   of the requestor by counting the MAGIC string args, and retrieve the type
   and client name accordingly.  Once inferred, ensure that the 'acks' sent
   to the requestor are in synch with the requestors expectations.  
--------------------------------------------------------------------------------
Etrack Incident = ET386568

Description:
   Schedules that specify "instant recovery" should not have backups that
   multi-stream.  It is possible that a policy may try to multi-stream in
   the following circumstances:
   - Frozen image enabled.
   - PFI enabled.
   - Multi-streaming enabled.
   - Policy contains schedules with instant recovery enabled.  
--------------------------------------------------------------------------------
Etrack Incident = ET340729

Description:
   If the bpgetmedia options -h and -M were both specified, the -h argument
   would always inherit the -M argument, and the original -h value would be
   lost.  
--------------------------------------------------------------------------------
Etrack Incident = ET323265

Description:
   bprd could not clean a vault session directory that contains files larger
   than 2 Gigabytes.  

Additional Notes:
   To resolve this issue, use lstat64() for LARGEFILES platforms in the
   recursive_dir_delete() function.  Used primarily by bprd for periodic
   deletion of log files (and vault session files)... and by bpdbm for
   removing expired back-ups.  
--------------------------------------------------------------------------------
Etrack Incident = ET314638 ET292805

Description:
   "bpdbm -consistency" failed to move out corrupted image files in
   "catstore".  It also failed to move out the zero byte image files
    (which should have been handled by cleanup but it also failed).  

Additional Notes:
   On UNIX platforms, "bpdbm consistency" failed to move out the corrupted
   image files if /usr/openv/netbackup/db and
   /usr/openv/netbackup/db/images/<client> were not on the same volume. In
   such case, the corrupted image files will be left to their original
   locations.  
--------------------------------------------------------------------------------
Etrack Incident = ET374767

Description:
   Acsd timeouts were increased to accommodate overloaded ACSLS server.
--------------------------------------------------------------------------------
Etrack Incident = ET392050

Description:
   Checkpoint-enabled user backups were performing an automatic retry
   (limited by the failure history). But user backups should not auto retry.  
--------------------------------------------------------------------------------
Etrack Incident = ET312346

Description:
   If you try to cat_convert an image where the policy name contains a
   Period (full stop), and the letter (f) follows immediately after that full
   stop, then the cat_convert says the file does not exist.  

Workaround:
   To avoid this issue, do not create policies with names that contain ".f".  
--------------------------------------------------------------------------------
Etrack Incident = ET303529 ET317435

Description:
   Backup jobs with multiple streams where the individual stream selections
   start with the same path can produce image files which, when browsed, do no
   list all of the files in the backup.
   
   Also, .f files created in NetBackup 4.5 versions for non-standard,
   non-Windows clients may not show all files in the backup when browsed
   under NetBackup 5.1.  
--------------------------------------------------------------------------------
Etrack Incident = ET312384

Description:
   Policy names that contain ".f" cause premature EOF on Full TIR Backups on
   the first try of jobid and when following an incremental.  

Workaround:
   To avoid this issue, do not use policy names that contain ".f".  
--------------------------------------------------------------------------------
Etrack Incident = ET352027

Description:
   If an attempt to create a binary .f file is done with file entries without
   user or group data, the attempt would fail but leave a zero-length .f
   file.  This caused a subsequent attempt to create the .f file in ASCII
   format.
--------------------------------------------------------------------------------
Etrack Incident = ET375769

Description:
   If many Flashbackup images are created using an ASCII image format,
   cleanup fails with the message, "too many open files".  

Workaround:
   To avoid this issue, use the binary .f file format to create the
   Flashbackup images.  
--------------------------------------------------------------------------------
Etrack Incident = ET374771 ET388368

Description:
   Tlhcd may dump core when running dump_tables from ltid -tables command.
   
   Tlhcd needs to check for a media ID with all spaces for an ID to be the
   last media in the list returned from the robot controller.  
--------------------------------------------------------------------------------
Etrack Incident = ET371650

Description:
   An XBSA application would hang during a shared memory backup (such as, the
   client and the media server are the same machine) if a problem occurred
   that caused the bptm to exit. The BSASendData() call did not return.  
--------------------------------------------------------------------------------
Etrack Incident = ET336233

Description:
   You no longer need to create the
   /usr/openv/netbackup/client/Linux/IBMzSeriesLinux2.4.21 directory prior to
   installing the patch if you plan to use this client type.  
--------------------------------------------------------------------------------
Etrack Incident = ET372549

Description:
   A stack trace showed that bpbrm was in the exit() system call, causing
   bpbrm to dump core.  This does not affect backups or restores.
--------------------------------------------------------------------------------
Etrack Incident = ET336220

Description:
   The removal of a copy of a backup image was not occurring when the copy
   expired.  This occurred when the operating system was Linux and the file
   system (where the NetBackup database resided) was VxFS.  
--------------------------------------------------------------------------------
Etrack Incident = ET336273

Description:
   Fixed the patch install to handle tar command options that were unique to
   Alpha.
--------------------------------------------------------------------------------
Etrack Incident = ET317430

Description:
   Bpbrm processes would hang after a backup was complete, or on Windows media
   servers, an unrelated bpbrm or bptm process would unexpectedly terminate
   with a 150 status (termination requested by administrator).
--------------------------------------------------------------------------------
Etrack Incident = ET341334

Description:
   If using Multiple Data Streams, there would possibly be a long delay in
   scheduling jobs.  The bpsched log would show many hosts_equal calls that
   were contributing to the delay.  
--------------------------------------------------------------------------------
Etrack Incident = ET313685

Description:
   The bpnbat -login text was wrong and has been corrected.
   
   
--------------------------------------------------------------------------------
Etrack Incident = ET394941

Description:
   User backups or archives that do not find an active policy/schedule to use
   were marked as resumable, but without the policy/schedule information, the
   job could not be resumed.
--------------------------------------------------------------------------------
Etrack Incident = ET394947

Description:
   A duplication may have been reported as successful if a defective tape
   drive would incorrectly report a tape mark on the read operation, followed
   by an I/O error.  This patch detects this error condition.
--------------------------------------------------------------------------------
Etrack Incident = ET203198 ET225277

Description:
   With NetBackup Explorer Extensions enabled, backup jobs started by using
   the "Backup Now" context-sensitive menu item will no longer fail with a
   status 12.  

Workaround:
   To avoid this issue, use NetBackup to do backups.
--------------------------------------------------------------------------------
Etrack Incident = ET371680

Description:
   The synthetic backup process can begin reading the wrong block number from
   tape during the synthetic backup image creation, causing the synthetic
   backup to fail.  This can happen when there are more than one tape drive
   is available to read the component images.
   
   The following sample log message is a symptom of this failure:
   
   bptm 12090 first XFERBLOCK indicated block number of 478341120 but initial
   positioning was for block 193352448
   
   This message indicating an unexpected block number read will be seen in
   both the job details for the synthetic backup job, and in the bptm log.  

Workaround:
   You can avoid this problem by downing drives so that only two drives
   are available for creating the synthetic backup - one drive for reading
   and one drive for writing.
   
   After the synthetic backup job actually mounts the first volume to be read
   and mounts the volume to be written, you can safely increase the number of
   up drives.
   
--------------------------------------------------------------------------------
Etrack Incident = ET270455

Description:
   Directories were not listed when browsing a backup on a policy that
   allowed multiple data streams .
--------------------------------------------------------------------------------
Etrack Incident = ET366161 ET366163

Description:
   In certain firewall configurations, multistream backups from Linux client
   may fail with status 25 if vnetd connect-back is used.
--------------------------------------------------------------------------------
Etrack Incident = ET311175 ET311173

Description:
   If the add_slave_on_clients command is unable to add the media server
   entries to one client, it may incorrectly update the media servers on
   subsequent clients.  Typically, only one media server will be added to
   the remaining clients.  

Workaround:
   To avoid this issue, use the Host Properties interface or the
   bpsetconfig admincmd to update the server list for the clients.
--------------------------------------------------------------------------------
Etrack Incident = ET300174 ET299670 ET300175 ET300176

Description:
   bpbrm would hang if a database agent backup failed in a certain way.
   The job would appear to be running in the Activity Monitor until an
   administrator terminated it.  
--------------------------------------------------------------------------------
Etrack Incident = ET322850 ET322817 ET322851

Description:
   When a server includes multiple network interfaces and is configured
   For NBAC, other NetBackup machines may reject connections from the server
   with authorization failures.  This patch insures that the server presents
   the correct credential to the other NetBackup machines.  
--------------------------------------------------------------------------------
Etrack Incident = ET326687

Description:
   bpbrm log messages were not being pulled in by backuptrace for multiplexed
   backups.
--------------------------------------------------------------------------------
Etrack Incident = ET376483

Description:
   A vulnerability has been confirmed in the NetBackup Volume Manager
   daemon (vmd).  By sending a specially crafted packet to the Volume
   Manager, a stack overflow occurs.  This is caused by improper bounds
   checking.  Exploitation does not require authentication, thereby
   allowing a remote attacker to take over the system or disrupt the
   backup capabilities.  Further testing and code inspection has revealed
   that all other NetBackup 5.0 daemons are potentially affected in the
   same manner.  
   
   Therefore, any Master Servers, Media Servers, Clients and Console
   Machines at this version level are subject to this vulnerability.  
   However, NetBackup 5.0 database agents are not affected by this issue.
   
   Refer to the Related Document section for more details.  
--------------------------------------------------------------------------------
Etrack Incident = ET378008

Description:
   The vmoprcmd -comment <drive_index> <comment_string> command would
   overwrite the serial number specified if the comment_string size was 26.
--------------------------------------------------------------------------------
Etrack Incident = ET396047

Description:
   The robotic inventory, vmupdate, may fail after encountering a media of
   different density. The problem is that vmupdate was not taking the media
   type of standalone media (if available) into consideration while
   determining the media type of the media to be moved into the robot.  
--------------------------------------------------------------------------------
Etrack Incident = ET401485

Description:
   If the forked re-read storage unit process detected a transient error, it
   would error, then exit and remove the scheduler message queues.  
--------------------------------------------------------------------------------
Etrack Incident = ET384759

Description:
   A potential race condition exists if a Windows client backup receives a
   134 status (resource busy).
   
   The race condition was caused because the bpfis process had already
   Started taking a snapshot and bpfis has no knowledge of the resource
   failure. So, the snap is taken, but when it is completed, there is no
   bpbrm and bpsched (child) process running to accept the snap information.
   
   When the job is later re-scheduled, this (unrequested) snap information
   Was passed back to the new bpbrm and bpsched.  The unrequested snap
   Information confused bpsched (child).  
--------------------------------------------------------------------------------
Etrack Incident = ET294366

Description:
   vmd would hand after 21 days under VxSS/NBAC.
--------------------------------------------------------------------------------
Etrack Incident = ET299436

Description:
   An extended value (90 days, for example) for the "Keep Logs for x Days"
   global configuration setting appears not to work.  The debug logs are
   still deleted after about a month on NetBackup media servers that
   are included as part of the catalog backup configuration.  
--------------------------------------------------------------------------------
Etrack Incident = ET404678

Description:
   Bprd was dumping core, and the stack trace looked similar to the following:
       
   Core was generated by `bprd -dontfork -mpxmain'.
   Program terminated with a signal 11, "Segmentation fault".
   
   #0  0x00034ccc in write_monitor_status ()
   (gdb) where
   #0  0x00034ccc in write_monitor_status ()
   #1  0x0003f204 in write_clnt_monitor_status ()
   #2  0x0004f7e8 in terminate_restore ()
   #3  0xfdd9efc8 in __tbl_10_small_start () from /usr/lib/libc.so.1
   #4  0xfdd53990 in posixgetdst () from /usr/lib/libc.so.1
   #5  0x0005f9ac in run_restores ()
   #6  0x00063bd4 in run_mpx_main_bprd ()
   #7  0x00021ad8 in main ()
--------------------------------------------------------------------------------
Etrack Incident = ET300998

Description:
   RMAN - alternate client restore times out after a successful partial
   restore by media server.
   
   Though this problem was encountered while doing an Oracle alternate client
   restore, the nature of the problem is such that it can happen for any
   XBSA-based partial restore.
   
   When the client is done restoring - partial restore - but there is still
   data to be read, the client sockets are closed but bptm/bpdm wait for the
   data to be read causing it to eventually time out with an error.  
--------------------------------------------------------------------------------
Etrack Incident = ET410097

Description:
   The snap driver needs to be compiled on solaris 10 to run on solaris 10.
--------------------------------------------------------------------------------

Etrack Incident = ET349769 ET419204

Description:
   Group or User names of longer then 31 characters would cause the catalog
   to consider an image invalid.  Invalid images are put in the db.corrupt
   directory, and the media is reclaimed.  This change enables the use of
   32-character-long user and group names, which is what bpbkar bpbrm
   supports.  When an image is found to be invalid the corresponding media
   was frozen to keep the media from being reused.  

Workaround:
   To avoid this issue, do not use user or group names longer then 31
   characters, and when cleanup is run, check for images that are put in
   the db.corrupt directory and import them if required.  
--------------------------------------------------------------------------------
Etrack Incident = ET343034

Description:
   Changes to fix "bplist returns duplicate listings for a DB2 image" caused
   queries of these images in ASCII mode to not work correctly.  This change
   makes DB2 work with both binary and ASCII.  

Workaround:
   To avoid this issue, use the binary catalog on servers that are backing up
   DB2 databases.  
--------------------------------------------------------------------------------
Etrack Incident = ET411553

Description:
   If using the command line bpubsora and bpubsdb2 would core dump
   immediately upon execution.
       
   If using the jbpSA interface, and the Oracle instance was clicked on, the
   application would immediately exit with the message, "bin sh:6491 abort
   status 134" and it would dump a core file.  
--------------------------------------------------------------------------------
Etrack Incident = ET324702 ET289704 ET324226

Description:
   The XBSA Agent PFI backups resulted in status 156 errors.  In this
   failure, the scheduler is receiving a file list, ‘start’(FI_FILE_START)
   and file list ‘end’(FI_FILE_END), however, it would not receive a file
   list ‘body’.  
--------------------------------------------------------------------------------
Etrack Incident = ET371927

Description:
   Because of security relates issues, the NetBackup 5.0 MP6 patch will be
   released with the 3.23.59 version of MYSQL server binary.
--------------------------------------------------------------------------------
Etrack Incident = ET415715

Description:
   Backups of Japanese language strings containing '[', '\', '*', '?' or '{' as
   a byte in a multibyte character fail.
--------------------------------------------------------------------------------
Etrack Incident = ET416208 ET414212

Description:
   A condition existed that resulted in excessive hosts_equal calls in the
   bptm log.  
--------------------------------------------------------------------------------
Etrack Incident = ET309015

Description:
   An issue existed that caused backups to fail when writing tir.  The
   bptm log contained the following information.
   
   15:46:57.554 [29981] <2> setup_tir_read: successfully connected to bpdbm to
   read TIR file, size is 577960 bytes + 0 GB
   15:46:57.554 [29981] <2> io_write_back_header: drive index 12,
   grzfs03_nafsp02_vol0_1108477090, file num = 50, mpx_headers = 0, copy 1
   15:46:57.658 [29981] <2> io_write_block: ndmp_tape_write_func returned 1024
   15:46:57.758 [29981] <2> NdmpMoverClient: ndmp_mover_set_record_size(64512)
   15:46:57.858 [29981] <2> NdmpMoverClient: ndmp_mover_set_window(0,
   18446744073709551615)
   15:46:58.058 [29981] <2> NdmpMoverClient: ndmp_mover_listen returned addr
   10.200.12.10 port 12091
   ------------
   15:50:42.667 [29981] <16> NdmpMoverClient: ERROR ConnectSocket - connect
   failed: 145 Connection timed out
   15:50:42.667 [29981] <16> NdmpMoverClient: ERROR ConnectSocket failed
   15:50:42.668 [29981] <16> NdmpMoverClient: ERROR Start failed
   15:50:42.947 [29981] <2> NdmpMoverClient: mover halted reason
   NDMP_MOVER_HALT_ABORTED

Additional Notes:
   To disable NdmpMoverClient in netbackup/db/config/ndmp.cfg, add the
   string, NDMP_MOVER_CLIENT_DISABLE.
--------------------------------------------------------------------------------
Etrack Incident = ET394971

Description:
   A restore from a FlashBackup image hung on a Linux server with an
   indication of a double free of a memory allocation.
--------------------------------------------------------------------------------
Etrack Incident = ET414056

Description:
   NDMP DAR restore fails when a tape has been imported
   or ‘FORCE_RESTORE_MEDIA_SERVER’ is set to a different NetBackup platform
    (UNIX to Windows or Windows to UNIX).  The restore fails with the
   following error:
   
   'NDMP_LOG_NORMAL 0 RESTORE: FILE_NAME: File does not exist in backup.'  
   
   

Workaround:
   To avoid this issue, use the same platform (UNIX or Windows) where the
   backup was performed.  

Additional Notes:
   The actual problem exists on the backup site. To do DAR restore using a
   different NetBackup platform (UNIX to Windows or Windows to UNIX), you
   have to perform a backup using this patch.  
--------------------------------------------------------------------------------
Etrack Incident = ET375811 ET236537

Description:
   When backing up and restoring large volumes with FlashBackup on windows
   various problems existed, these included memory allocation problems for
   incremental backups due to the fact that Microsoft does not guarantee
   contiguous free space.  Also there were various 32 bit integer overflow
   conditions in the sfr, single file restore, code.  
--------------------------------------------------------------------------------
Etrack Incident = ET393136 ET414942

Description:
   Added the following robots:
   - Certance CLL6400
   - Falcon Stor VTL
   - HP 1x8 Autoloader
   - NEC T16A
   - SONY CSM-20
   - STK VDRIVE
   
   Updated the following robots:
   - ADIC Scalar 24
   - ADIC 224
   - ADIC 234
   - ADIC 448
   - ADIC 458
   - ADIC Scalar 24
   - ADIC Scalar 1000
   - ATL M1500
   - ATL M2500
   - ATL P3000
   - Compaq MSL5000
   - Dell PV-132T
   - Exabyte VXA-2 1x10 1U
   - Exabyte 440
   - Exabyte 480
   - HP MSL5000
   - HP MSL6000
   - HP VLS
   - IBM 3583
   - IBM 7337
   - Spectralogic PYTHON

Additional Notes:
   Devices can be included in this file before they are officially supported.
--------------------------------------------------------------------------------
Etrack Incident = ET417526

Description:
   Security permissions of some of the temp files were set to 666.  Those
   Permissions have been changed to 644.  
--------------------------------------------------------------------------------
Etrack Incident = ET422757 ET499607

Description:
   Synthetic Backup produces an image that contains only the last
   incremental if a previous full synthetic backup (from incremental
   backups) was manually expired.  
--------------------------------------------------------------------------------
Etrack Incident = ET416687

Description:
   If a customer created an image list (bpimage -create_image_list -client
   <client) after an incremental TIR backup and then ran a full or cumulative
   differential backup, the TIR_INFO field in the image was not updated to 10.
   As a result, TIR information was not pruned from the backup image.  

Workaround:
   To resolve this issue, remove image list files from the client directory.
--------------------------------------------------------------------------------
Etrack Incident = ET419090

Description:
   Loop-back file systems were being backed up as local file systems causing
   the data to be backed up twice.
--------------------------------------------------------------------------------
Etrack Incident = ET422894

Description:
   Staging from a disk staging storage unit that contained a space would
   result in images that reside on the staging storage unit to not be
   duplicated properly. However, the staging job would appear to succeed in
   the Activity Monitor.

Workaround:
   To avoid this issue, duplicate images to a new DSSU that do not contain a
   space, and change any policies that write to original DSSU to use the new
   DSSU.
--------------------------------------------------------------------------------
Etrack Incident = ET342808

Description:
   When bpbrm encountered an error during backup, bpbkar did not shut down
   properly. On HP-UX 11.11 platforms, it would get stuck in an infinite loop
   until an alarm signal went off after 300 seconds, causing millions of debug
   messages to be written to the debug log. On other UNIX platforms, bpbkar
   shuts down abnormally because of a segmentation fault.

Additional Notes:
   The new Internationalization message is:
   
   Error reading filelist from server, name_len = %d, errstatus = %d
--------------------------------------------------------------------------------
Etrack Incident = ET355631

Description:
   When bpbackup is run with the option to wait for the completion of the
   backup job, a firewall between bpbackup and bprd can shut down the socket
   while bpbackup waits for the completion status.  This causes backint to
   get stuck in an infinite loop during database backups.  
--------------------------------------------------------------------------------
Etrack Incident = ET423634

Description:
   A change has been made to avert a potential vulnerability in a Java
   authentication service that runs on VERITAS NetBackup servers and clients.
   This change prohibits remote attackers from executing arbitrary code on a
   targeted system.  In addition, Symantec recommends that users block the
   affected ports from external network access.
--------------------------------------------------------------------------------
Etrack Incident = ET421519

Description:
   The first line of the header of the bpclimagelist command was occasionally
   not displayed.  This potentially could happen for any server type, but was
   seen when the master server was Linux.  This would affect BMR processing.  
--------------------------------------------------------------------------------
Etrack Incident = ET413511

Description:
   For a policy that uses wildcards (such as, *) in the file list and enables
   multiple data streams, NetBackup will not include any dot files (such as,
   .rhosts) at that directory level in the backup.
--------------------------------------------------------------------------------
Etrack Incident = ET321343

Description:
   Bpbkar debug logs are currently limited to no more than 2 Gigabytes in
   size.  This change enables those logs to grow beyond 2 Gigabytes.
--------------------------------------------------------------------------------
Etrack Incident = ET427127

Description:
   visd consistently went down on HP_UX platforms when using the new 3.23.59
   nbdbd binary.
--------------------------------------------------------------------------------
Etrack Incident = ET420747

Description:
   The disks on Hitachi arrays at higher ports cannot be backed up when
   using Shadowimage, because the CSH and Port number on the disks are
   more than two digits long.
   
   The fix enables ShadowImage to backup Tagma array disks that reside on
   upper ports.  
--------------------------------------------------------------------------------
Etrack Incident = ET429022

Description:
   A change was made to ensure the SCSI inquiry output for Hitachi array
   from pagecode83 always to be 4 digits so that shadowimage will work.
   
   In rare cases, SCSI inquiry output for Hitachi array from page code 83
   (bytes 32 and 33) returned two digits instead of four digits of numbers.
   
   This fix filled in the digits and enabled Shadowimage to work.  
--------------------------------------------------------------------------------
Etrack Incident = ET430821

Description:
   Binding to a socket that was already in use caused the following error to
   appear in the bpsd log file:
   
   10061 - WSAECONNREFUSED

Workaround:
   To avoid this issue, use vnetd.
--------------------------------------------------------------------------------
Etrack Incident = ET426765

Description:
   User initiated backups fail while creating .f file.
   The policies were created on the fly.  
--------------------------------------------------------------------------------
Etrack Incident = ET498524

Description:
   In a rare condition one or more backups would be corrupt when using Inline
   Tape Copy (ITC) with multiplexed backups.  This would potentially happen
   under the following conditions:
   
   - ITC was enabled with 2 or more active copies.
   - Multiplexing was enabled with 2 or more active backups.
   - The schedule configuration had "continue if the copy fails" flag
   set for the copy that failed in the next step.
   - A copy other than the first active copy encountered "end of media", so
     a new media was needed for this copy.  Then, there was a problem getting
     or setting up media (for example, writing the media header) such that
     this copy failed before continuing the backup.
   - The last buffer written to the first active copy prior to the "end of
   media" on the other copy was not the first active backup.
   

Workaround:
   To avoid this problem, do one of the following:
   
   - Edit the schedule configuration to disable the "continue if copy fails"
   flag.
   - Disable multiplexing.
   - Disable ITC.
   
================================================================================



=========
NB_50_5_M
=========
Etrack Incident = ET269449

Description:
   This version of the SG driver adds support for Solaris 10.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET258456

Description:
   Added support for the following drive types in ACS robots:
       STK 10000A (T1)
       HP LTO-3
   
   Added support for the following media types in ACS robots:
       T10000T1
       T10000TS
       T10000CT
       LTO-400G
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET271527

Description:
   Linux added three file system types (tmpfs, usbdevfs, and devpts) that are
   not useful to backup.  A change has been added to automatically skip them
   on backups.

Workaround:
   To avoid this issue, the file systems have been added to the exclude list.

   (NetBackup Clients:  Linux2.4 )
--------------------------------------------------------------------------------
Etrack Incident = ET276832

Description:
   User archive or synthetic backups that have checkpoint enabled and
   receive a status 134 (resource busy) will be rescheduled. But the Activity
   Monitor would list the job as "done" with a status 134.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET277275

Description:
   A potential memory leak in the NetBackup scheduler has been resolved.

Additional Notes:
   This patch contained level src/nb/sched/dbmisc.c 1.88.4.22.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET277276

Description:
   Corrected a situation that caused the scheduler to lose a backup in the
   middle of a catalog backup.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET276090

Description:
   Added 64-bit Oracle support for linux on AMD64 and EM64T platforms.

   (All UNIX NetBackup Servers and NetBackup Clients:  Linux2.4)
--------------------------------------------------------------------------------
Etrack Incident = ET270620 ET270622

Description:
   There are cases when a NetBackup administrator can administrate most things
   on a NetBackup master server (such as policy configuration), but is no
   allowed to access Host Properties on clients.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET279000

Description:
   Corrected an issue with the Sony TSL-11000 tape library that caused an error
   when moving the cleaning tape to its appropriate slot.

Workaround:
   The cleaning actual works, but the cleaning statistics do not get updated.  
   Running a tpclean -M <drive name> after the failed clean corrects this issue.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET283487

Description:
   Initiating multiple Oracle backups was causing a crash to occur in the
   scheduler. Each of the Oracle backups attempted to read/update/write a
   group ID file (used for synching snap backups), but there was no lock
   mechanism on the file access, so each Oracle backup had the potential to
   corrupt another?s file access.

Workaround:
   To avoid this issue, you should stagger window openings for Oracle backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET276438

Description:
   xinetd on a SuSE machine has a different path than that on a RedHat machine.
   Three scripts were changed to choose the proper path depending on the
   Linux Operating System installed on the client.

   (NetBackup Clients:  Linux2.4 )
--------------------------------------------------------------------------------
Etrack Incident = ET272607

Description:
   If a set of configuration options in bp.conf is very large, it could cause
   bpgetconfig and the GUI to not display any of the items.  This applied to
   items in bp.conf with the same name, for example CONNECT_OPTIONS or SERVER.
   In addition, if that set of configuration options was edited in the GUI,
   for example to add a new server to the CONNECT_OPTIONS, all of the
   previous data for that set of options was lost and only the new data would
   be in bp.conf.

Workaround:
   To avoid this issue, reduce the number of configuration options with
   the same name.

   All NetBackup Windows Clients All NetBackup PC Clients All NetBackup UNIX
   Clients
--------------------------------------------------------------------------------
Etrack Incident = ET272607

Description:
   If a set of configuration options in bp.conf is very large, it could cause
   bpgetconfig and the GUI to not display any of the items.  This applied to
   items in bp.conf with the same name, for example CONNECT_OPTIONS or SERVER.
   In addition, if that set of configuration options was edited in the GUI,
   for example to add a new server to the CONNECT_OPTIONS, all of the
   previous data for that set of options was lost and only the new data would
   be in bp.conf.

Workaround:
   To avoid this issue, reduce the number of configuration options with
   the same name.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET262341

Description:
   Legacy Encryption may fail if the key file on the client was large.  For
   example, if 39 or more pass phrases were added to a key file on a Solaris
   client, encrypted backups or restores were likely to fail.  The error
   message "bpbrm - read() or KEYREADFD failed" would have typically been
   reported.

Workaround:
   To avoid this issue, delete some pass phrases from the key file if they
   are no longer needed.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Etrack Incident = ET284895

Description:
   When running a Restore job, the information about the job did not appear in
   the Job Details -> Job Overview tab in the Activity Monitor.  However, the
   job would still complete successfully.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET286714

Description:
   The bpimmedia -spangroups output was not displaying a span group if the
   span occurred in a copy other than copy 1, and one or more preceding
   copies were expired.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET288891

Description:
   The following message would appear in the syslog on Linux 9+ systems when
   running a stand-alone bpcd:
   
   kernel: application bug: bpcd(1801) has SIGCHLD set to SIG_IGN but calls
   wait().

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Etrack Incident = ET289147

Description:
   Corrected an issue where backups were possibly skipped if the frequency was
   a multiple of 24-hours (interval mode) and the backup window was greater
   than 24-hours. (Only the first backup was run in the failure scenario.)

Workaround:
   To avoid this issue, make your backup windows no greater than 24-hours.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET260839

Description:
   Backing up files which do not have a group name would fail with a Status 13.
   

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Etrack Incident = ET285171

Description:
   A Synthetic Backup Image could be expired while it was still being written.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET277846

Description:
   Staging of images that were no longer in the catalog caused a 190 error.

Workaround:
   To avoid this issue, manually delete the images that are not in the
   catalog.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET277814

Description:
   Restores with a Hitachi NAS filer failed. The NDMP log showed the
   NDMP_DATA_START_RECOVER has the restore path, and listed it twice.

Workaround:
   Restores will work as long as the file-system path (the path that was
   initially backed up) is not selected. If the entire image is needed for
   restore, select all the files and directories in the file-system path
   instead.

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET287624

Description:
   When a client is included in multiple policies, synthetic backups may fail
   if incremental backups for these policies were initiated at the same time.

Workaround:
   To avoid this issue, set the "Maximum jobs per client" master server
   property to 1.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET292764 ET289149

Description:
   Corrected an issue where backups were possibly skipped if the frequency
   was a multiple of 24-hours (interval mode) and the backup window was greater
   than 24-hours. (Only the first backup was run in the failure scenario.)

Workaround:
   To avoid this issue, make the backup windows no greater than 24-hours.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET293145 ET219875

Description:
   NetBackup processes may core dump if an error occurs. The stack from the
   core file showed problems with the ndmp_user_scsi() or ndmp_user_tape()
   functions.

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET277782

Description:
   bpdm would corrupt a TIR file when a "disk full" condition was encountered.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET293094

Description:
   When the sui bit is set on some admin commands on AIX 5.2, the system
   oslevel command will leave a directory in /tmp.  This does not happen for
   the root user.

   (All NetBackup Servers: RS6000)
--------------------------------------------------------------------------------
Etrack Incident = ET268043

Description:
   When a local file system is auto-mounted to itself, NetBackup can
   incorrectly skip the backup of the local file system on SGI systems.  This
   occurs when the hostname returned from the uname command and the hostname
   stored in /etc/mtab are not both fully-qualified or both short (for
   example, uname returns "machine" and /etc/mtab contains
   "machine.domain.com").

Workaround:
   To avoid this issue, make sure the hostname returned by the uname command
   and the hostname specified in /etc/mtab both use either the
   fully-qualified hostname or the short name.

   (NetBackup Clients:  SGI65 )
--------------------------------------------------------------------------------
Etrack Incident =  ET287111

Description:
   An error occurred when a SAP online backup was started and a Tablespace
   (PSAPES46CI) in the Oracle DB was placed in BACKUP MODE by the agent. The
   backup received a STATUS 134 error in BPTM & BPBRM. The bpbkar, bptm, and
   bpbrm processes would exit properly, however the SAP agent process
   remained active, waiting for the backup to be queued again and rerun. This
   caused the Tablespace (PSAPES46CI ) to be left in BACKUP MODE.
   
   The following is an example log of how this error occurred.
   
   03:08:23.624 [25925] <4> backint_monitor_log: [25926.00] 03:08:16 INF -
   Processing /oracle/P01/sapdata4/es46ci_1/es46ci.data1
   
   03:08:23.624 [25925] <4> backint_monitor_log: [25926.00] 03:08:19 INF -
   BRSAP BR280I Time stamp 2004-11-28 03.08.19
   
   03:08:23.624 [25925] <4> backint_monitor_log: [25926.00] 03:08:19 ERR -
   bpbkar killed by SIGPIPE                                    
   
   Bpbkar gets 40 error, bptm & bpbrm 134 error
   The backup gets requeued in the scheduler and then rerun. Unfortunately on
   the Client the backup fails with this error:
   
   03:08:23.624 [25925] <4> backint_monitor_log: [25926.00] 03:08:19 INF -
   Waiting for mount of media id PFC121 on server S01Q02 for writing.
   
   03:08:53.623 [25925] <4> backint_monitor_log: [25926.00] 03:08:46 INF -
   Waiting in NetBackup scheduler work queue on server s01a20
   
   03:09:23.623 [25925] <4> backint_monitor_log: [25926.00] 03:09:18 INF -
   Waiting in NetBackup scheduler work queue on server s01a20
   
   03:09:53.622 [25925] <4> backint_monitor_log: [25926.00] 03:09:51 INF -
   Waiting in NetBackup scheduler work queue on server s01a20
   
   03:10:08.622 [25925] <4> backint_monitor_log: [25926.00] 03:09:57 INF -
   Processing /oracle/P01/sapdata4/es46ci_1/es46ci.data1
   
   03:10:08.622 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 INF -
   BRSAP BR280I Time stamp 2004-11-28 03.09.59
   
   03:10:08.622 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 INF -
   BRSAP #BEGIN /oracle/P01/sapdata4/es46ci_1/es46ci.data1
   
   03:10:08.622 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 INF -
   BRSAP BR280I Time stamp 2004-11-28 03.09.59
   
   03:10:08.623 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 INF -
   BRSAP BR345E Error in backup utility while switching the status of
   file /oracle/P01/sapdata4/es46ci_1/es46ci.data1
   
   03:10:08.623 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 INF -
   BRSAP BR280I Time stamp 2004-11-28 03.09.59
   03:10:08.623 [25925] <4> backint_monitor_log: [25926.00] 03:10:00 ERR -
   SAP begin alter table space failed
   
   Because the Tablespace (PSAPES46CI) is still in BACKUP MODE from the first
   backup attempt.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Etrack Incident = ET286655

Description:
   An error occurred that would cause bpverify to fail with a 191 error on a
   single master/media server.  The operating system would time out the
   socket before the processing finished.
   
   CLIENT_READ_TIMEOUT = 1800
   tcp_keepalive_interval = 7200000
   tcp_time_wait_interval = 1000
   
   06:48:42.943 [7244] <2> get_adaptable_string: (4) network read() error:
   Connection reset by peer (131)
   06:48:42.944 [7244] <2> db_FLISTreceive: expected 9, got 1

Workaround:
   To avoid this issue, set the default time-out on sockets to be longer, or
   perform a larger number of small backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET279260 ET293080

Description:
   Corrected maislot timing issues that sporadically occur on certain robots
   when using AUTO_UPDATE_ROBOT.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET295112

Description:
   The scheduler has been modified to correctly process a status 134 error
   race condition in a multiplex backup.
   
   This condition occurs when a multiplex child bpsched starts a backup, but
   the storage unit is over committed. A status 134 is sent back to the
   multiplex child bpsched, but the child sched is already adding another
   backup to the multiplex group.
   
   The child sched attempts to send the new jobs information to bpbrm, but
   bpbrm has exited and the connection is defunct. The scheduler was
   incorrectly logging the defunct connection in the job progress log. With
   this change scheduler will not log the broken socket, but will correctly
   read and process the 134 from the inbound connection.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET294712

Description:
   bpretlevel loops were echoing the phrase 'What?' when changing retention
   levels.  No input seems to be a satisfactory response to exit the loop
   until a ctrl-C is entered.  There did not seem to be any way to change
   retention levels with the client interface because of this bug.  A
   correction within the source code has orrected this problem.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET287371

Description:
   When multiple disk staging jobs are initiated and the DSSU is at or
   near capacity, some jobs may fail with a status 84 error, "disk full,
   attempting to remove the same staged images".
   
   The BPDM logs contained entries that indicated no matching entries were
   found in the catalog, (db_IMAGE: db_end_sts() failed: no entity was found).

Workaround:
   If the DSSU is nearing capacity, manually remove the previously staged
   images. These images are identified by an additional zero-length file of
   the same name, with a ".ds" extension.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET296132 ET270045 ET268600 ET260459 ET273679 ET220083
ET260459 ET274370

Description:
   Added support for the following libraries:
   
   - ARCHIVE Python 06241-XXX910B
   - Certance CLL3200
   - Diligent VTF Open
   - Exabyte VXA 1x10 Autoloader
   - Exabyte 1x7 Autoloader
   - IBM DDS4 Autoloader
   - IBM 3623 2SX
   - SEAGATE DAT 9SP40-400910B"
   
   Added support for the following drives:
   
   - Exabyte VXA-2
   - HP LTO3
   - HP VS160
   - IBM Ultirum-TD3, 3580-TD3
   
   Changed Serialization for the following libraries:
   
   - Sony CSM-200, CSM-100, CSM-60

Additional Notes:
   Devices can be included in this file before they officially supported.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET296424

Description:
   This bpbrm binary fixes an intermittent socket problem that occurs during
   database backups. The backup may fail with a status of 54.
   
   Examination of the bpbrm debug log will show the following:
   
   07:10:52.662 [3832.2768] < 16> bpbrm listen_for_client: listen for client
   timeout during accept from name listen socket after 60 seconds
   
   Examination of the dbclient debug log will show the following type of
   message:
   
   02:24:07.956 [528.660] < 8> connectSock: WRN - connect() to server failed,
   No connection could be made because the target machine actively refused it.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET274191

Description:
   The VxFS team is changing their implementation of ACLs on VxFS file
   systems on the AIX platform.  NetBackup has changed to handle this new
   implementation.

   (NetBackup Clients:  RS6000_433 RS6000_51 )
--------------------------------------------------------------------------------
Etrack Incident = ET297729

Description:
   Oracle BLI backups fail on Solaris 10 when trying to backup multiple VxFS
   partitions.  For a second or later file system, the checkpoint mount may
   fail.  The bpbkar log would contain a message similar to the following:
   
   11:58:55.199 [8353] <4> bpbkar set_clone_context: /usr/sbin/mount -F vxfs -
   o
   ckpt=NetBackup_full_orac101_1106243854
   /dev/dsk/c1t0d0s4:NetBackup_full_orac101_1106243854
   /tmp/NetBackup_checkpoints/NetBackup_full_orac101_1106243854 >/dev/null
   2>/dev/null
   11:58:55.464 [8353] <16> bpbkar set_clone_context: ERR - Cannot mount
   checkpoint NetBackup_full_orac101_1106243854 on
   /tmp/NetBackup_checkpoints/NetBackup_full_orac101_1106243854.

   (NetBackup Clients:  HP RS6000_433 Solaris Solaris9 )
--------------------------------------------------------------------------------
Etrack Incident = ET287624

Description:
   When simultaneous differential incremental backups are done for the
   same client, a subsequent synthetic backup for that client may fail.

Workaround:
   To avoid this issue, do not run simultaneous backups on the same client.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET272781

Description:
   Unable to install Java patch from a read-only location because of script
   renaming that takes place during the install.  Java patch should come
   packaged with scripts appropriately named so that script renaming does not
   need to occur during the installation. This change is also accompanied by
   one to tar_pack.pl which creates the package.
--------------------------------------------------------------------------------
Etrack Incident = ET272979

Description:
   During restore 'bprd -mpxmain' is trying find a VxPSP library in the path
   it was built in. The workaround is to do the following post-patch install
   on HP-UX:
   
   ln -s libvxicuucST.sl libvxicuucST.0
   ln -s libvxicudataST.sl libvxicudataST.0
--------------------------------------------------------------------------------
Etrack Incident = ET301487

Description:
   
   Run sg.install to install the new sg driver post installation.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET300675

Description:
   If a /dev/rdsk device is used as the extended copy command target, the code
   to convert the disk path to the sg path for scsi-pass through commands
   fails if the disk is managed by the ssd driver. This will cause an error
   to be logged in the bptm log similar to the following:
   
   07:00:44.807 [13213] <2> setup_mover_tpc: open of passthru path
   /dev/rdsk/c1t50060E80034FED11d0s2 failed, unable to open
   /devices/pci@1f,0/pci@5/SUNW,qlc@1/fp@0,0/sgd@w50060e80034fed11,0:raw, No
   such file or directory

Workaround:
   To avoid this issue, configure the corresponding sg path directly in
   mover.conf

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET300678

Description:
   FlashBackup on Windows restores of compressed files were aborting when
   running on Solaris master servers.  The tar headers were being constructed
   incorrectly because of the handling of 64-bit values.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET302233

Description:
   Robot inventory update was not taking into account the media type for
   existing standalone media that have been moved into a robot. It was only
   looking at the barcode rules. This was the cause of the error "media
   type and volume group mismatch (101)". Robot inventory update now updates
   the media residences to be robotic without changing the media type.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET302340

Description:
   Scheduler was not preferring the local storage unit when using storage
   unit groups and in-line tape copy.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET303448

Description:
   Backups would fail with a disk full error (status 40) even though there was
   ample room on the disk.

Workaround:
   The problem was the result of the settings in /etc/security/limits. You can
   increase the file size value to eliminate this problem.

   (All NetBackup Servers: RS6000)
--------------------------------------------------------------------------------
Etrack Incident = ET303563 ET302347

Description:
   Main sched failed to recognize a media server that was initially off line,
   and then brought on line after main sched went active.
   
   If the media server was brought back while scheduler was not active or the
   master server was bounced, then everything worked correctly.
   
   This fix enables scheduler to ?pick up? the newly introduced media server
   using the /bp/bin/admincmd/bpschedreq ?read_stunits command.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET304467

Description:
   Staging/Duplication from a disk staging unit was not working when more
   than 32765 files were in the staging area.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET259303

Description:
   Enhancements have been added to bpdm and bpdbm to clean up fragments that
   used to be left on the media server when bpdm failures occurred.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET283834

Description:
   Under some circumstances, bpSALinfo was not able to obtain a lun value for
   a device from the SAL database, resulting in the entry not being updated
   with the WWN and lun value.

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET309222

Description:
   Backups were taking too long between job complete and the 'end writing'.
   Corrects were made to a recursive call in searching for a storage unit.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET299873

Description:
   Bptm, bpdm, or bpdbm occasionally crashed during TIR incremental backups.  
   Bprd, bpbrm, or bpsched might have been affected as well.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET299874

Description:
   Multiplexed backup jobs would occasionally hang.  This would happen if the
   bptm or bpdm parent process crashed, or if bptm could not successfully add
   the TIR fragment for a TIR backup.  It may have also happen due to a timing
   problem in bpbrm when new jobs were started.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET313368

Description:
   The fix to allow greater than 32K entries to be staged would only work if
   the DSSU and destination storage unit are on the same media server.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Etrack Incident = ET313395

Description:
   On Solaris 10 platforms, NetBackup now skips the backup of the new file
   system types ctfs and objfs.

Workaround:
   Add the mount points of any objfs and ctfs file systems into the exclude
   list on that client, for example, /system/object and /system/contract.

   (NetBackup Clients:  Solaris9 )
================================================================================



=========
NB_50_4_M
=========

Description:
   Resuming a checkpoint backup could result in the selection of a different
   media server if a failure occurred on the initial backup attempt.
   
   The failure scenario would look something like the following:
     1) Backup reached a checkpoint.
     2) Then failed (in this case a 13 "read failed" was received).
     3) The Scheduler attempts to programmatically re-submit.
     4) However the drives were re-set & downed.
     5) The result was that the job exited with a 213 "no storage units
        available" error.
   
   The original storage unit information was "lost" so the job could
   potentially resume on a different media server. If the job was resumed in
   this way, the backed-up data would reside on two different media servers
   and any subsequent restore would fail.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added Solaris10 support to the client drop-down list for this release.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   After an NDMP import occurred, the file count would drop by a very large
   number.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A backup with TIR and CPR enabled in the policy attributes and files,
   instead of directories specified in the policy file list, can fail with a
   status of 130 (signal 11 in the bpbkar log) if the backup continues for
   more than the checkpoint interval.

Workaround:
   Do not enter files in the policy file list, with TIR and CPR enabled in the
   policy attributes. TIR works on directories; therefore, specifying files in
   the policy file list does not accomplish anything.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   There is a potential for missed files if all the following conditions
   exist:
    - Advanced client backup
    - Incremental backup
    - Multi-stream backup
    - Backup stream is not stream 1
   
   In addition, files modified after the snapshot is taken but before the
   (non-stream 1) image is created may be missed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   "Ghost" backup jobs were appearing in the activity monitor.  For example, a
   deleted backup job could (partially) re-appear in the activity monitor.
   This would only happen to a deleted job that was one number lower than a
   multi-copy backup.
   
   And example of this scenario is:
     1) Job n finishes.
     2) Job n is deleted from the activity monitor.
     3) Multi-copy job n+1, n+2, n+3 finishes.
     4) Activity monitor incorrectly updates jobs n, n+1, n+2.
     5) Deleted job n (partially) re-appears.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When "Browsing for restore", files that ended in an asterisk (*) or an
   angstrom (@) would have that particular symbol (* or @) truncated. This
   prevented those files from being restored.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   If an in-line copy backup is multiplexed, and one of the copies gets a
   failure (but continues), then subsequent additions to this troubled
   multiplexed group generates excessive failure messages.
   
   This code change inhibits adding new backup jobs to multiplexed groups
   that contain a failing copy.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The bpplschedrec -deldayomonth command deletes a range of days instead of
   the specified day.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The scheduler does not automatically start "bpexpdate -deassignempty" to
   deassign media that do not have associated images after "bprecover -l" has
   been run.

Workaround:
   If a bprecover -l has been executed, remove the following file:
   
   UNIX:
   /usr/openv/netbackup/bin/deassign_lock
   
   Windows:
   VERITAS\NetBackup\Bin\deassign_lock
   
   Do not remove this file if a catalog recover (bprecover -r) has been done.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The cancel_all_exceeded_max_wait() function does not scale well in large
   environments. This change modifies the way incomplete jobs are processed
   when they have exceeded the maximum incomplete time.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The problem existed when the following conditions were true:
    - Snap backup (advanced client, Windows Open File Backup (WOFB)).
    - Incremental backup.
    - Multi-stream backup.
    - Stream 1 is not due, but is force started to perform the snap shot
   operation. Aggravated by ?frequency? mode (frequency is NOT multiple of
   24-hours).
   
   The problem can be minimized/eliminated by configuring for an all stream
   backup:
    - Use calendar based scheduling.
    - Use ?interval? mode (frequency is multiple of 24-hrs).
    - Avoid multi-streaming snap backups.
   
   This problem resulted because of an oversight in cloning a due stream into
   a non-due stream1. Remember, stream1 is not due, but is required to
   perform the snap shot operation. The due stream information is copied in
   mass to create the stream1 information, and then adjusted to make it
   stream1 unique. The error occurs because the backup times (last backup,
   last full backup, last incremental backup) were not updated to the correct
   stream1 values. As a result of this, stream1 uses the stream2 image
   creation times. Image creation times are used to compute the delta time
   which determines the targeted incremental backup files.
   
   Unfortunately stream2 also suffers the same failure. The stream2
   incremental backup should be based upon the stream1 snap time, but stream1
   was using the incorrect time for the snap. So, all the due streams suffered
   this failure.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Under some circumstances, an image file might be expired on the Master
   server and remain on the Media server. When media full is detected for a
   DiskStaging storage unit, the media server attempts to remove the two
   oldest images. If these images were no longer in the catalog on the Master
   server, they could not be removed and the backup would fail.
   
   The code was changed to remove the image files if the Master server no
   longer had them in the catalog.

Workaround:
   To avoid this issue, manually delete the image files that are causing the
   problems along with their .ds file.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A loss of multiplexed groups within bpsched.exe resulted in the failure to
   cancel multiplexed backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   VxFS_Checkpoint with PFI file promotion of hardlink creates two files.  

   (NetBackup Clients:  HP Solaris Solaris9 )
--------------------------------------------------------------------------------
Description:
   Disk Staging would not correctly relocate backup images from clients when
   a client name contained an "_" character.

Workaround:
   To avoid this issue, remove the "_" from the client computer names.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NetBackup Administration Console (on re-launch) and admincmd/bpschedule
   display incorrect or no values for specific_include_dates specified in
   calendar-based schedule for DSSU relocation schedule. However, this does
   not affect the configured schedule getting triggered at correct time. This
   behavior is observed only on HP-UX platform.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A race condition existed when two restore jobs needed the same piece of
   media, from the same server, when this media was not already in the media
   database.  This condition could cause the media to be added twice, and one
   of the restores to fail.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   In one instance, some catalog files were corrupted when the system ran out
   of disk space. This caused bpdbm to crash whenever it tried to load the
   corrupted catalog files. The command "bpdbm -consistency 2" also crashed
   every time these corrupted files were checked.
   
   The cause of this problem was the buffer overrun in libcatalog when loading
   corrupted files: the pointers and string lengths are wrong after mapping
   the file into memory. To address this problem, code was added to safe guard
   buffer overrun to fix the crash of bpdbm.  In addition, code was added in
   image.c to move corrupted catalog files out of /db/images (into /db.corrupt
   directory) when running "bpdbm -consistency 2".
   
   This new code resulted in the following new feature:
   
   "bpdbm -consistency" will move corrupted catalog files out of the catalog
   repository "/netbackup/db/images" and into "/netbackup/db.corrupt"
   directory.
   
   After running the above command, users will need to do the following:
   1. Record all file paths in /db.corrupt;
   2. Re-import the files in step 1;
   3. Delete files in /db.corrupt.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Improved VMD debug logging to better enable analysis for slow servers.
   
   Allow control of VMD child timeout, such that slow servers can be
   interrupted, before VMD encounters connection timeout's on other
   connections.
   
   Add "VMD_CHILD_TIMEOUT = xx", to the vm.conf file.  xx is the number of
   seconds to wait for a child to release its lock on the volume database.
   Allowable time is 60 to 360 seconds, where 360 is the default (a value
   used prior to this patch).

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bpbrm received a SIGSEGV and core dumped in the malloc system call. This
   issue has been resolved.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If the following conditions are met, the activity monitor will report
   a "status 50" failure for the child duplication jobs.  
   
   - A customer is configured with a disk-staging storage unit.
   - The DSSU's "Disk Staging Schedule" is configured to produce multiple
     copies.
   - A DSSU duplication takes longer than 20 minutes.
   
   NOTE: The single parent duplication job, as well as the "__DSSU_POLICY_"
   backup job, remains active in the activity monitor.  In addition, ALL
   processes remain active.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Following messages did not match:
   
   NecErr0363N "recveived an error from bptm request to suspend media"
   TrbErr0631N "received an error from bptm request to suspend media"
   
   NecErr0364N "recveived an error from bptm request to un-suspend media"
   TrbErr0632N "received an error from bptm request to un-suspend media"
   
   NecErr0378N "unable to get address of the local listen socket"
   TrbErr0646N "unable to get the address of the local listen socket"

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   On Internationalized platforms, the output of the usage statement would be
   incorrectly displayed in English.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   In certain configurations, calls to gethostbyname (3nsl) may fail. In such
   cases, VERITAS did not force a DNS lookup for a fully qualified host name.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   vmphyinv is unable to read the tape header in NDMP drives. This affects
   NDMP drives and the NetWare media server (that are handled using NDMP
   protocol).

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If NetBackup is configured to duplicate an image and also configured to
   change the retention level of the new copy to a retention level that is
   configured as "expires immediately", then the original image also appears
   to be expired at the end of the duplication. Any backup taken after that,
   returns the original image.

Workaround:
   Any subsequent backup would expunge expired copy from the image file, and
   bring back the original image (to image list and for restore operation).
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The Oracle recovery template would cause a core dump if it was used under
   the Template Administration window.  

Workaround:
   Place the Oracle/DB2 recovery templates in a directory other than where
   backup templates reside.
   
   Template administration looks at backup templates in the following
   directory :
   
   /usr/openv/netbackup/db/dbtemplates/oracle (Oracle - UNIX)
   /usr/openv/netbackup/db/dbtemplates/db2 (DB2 - UNIX)
   
   VERITAS\NetBackup\db\dbtemplates\oracle (Oracle - Windows)
   VERITAS\NetBackup\db\dbtemplates\db2 (DB2 - Windows)

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   During an alternate client restore of Informix, the physical restore is
   successful. However, the logical log restore always fails while restoring
   the last log requested. This happens whether the restore is a full
   restore, point-in-time restore or point-in-log restore.
   
   The transaction appears to be complete on the client and the client
   disconnects the socket connections to the media server. The media server
   would not expect this and would exit with status 24.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   SCSI-3 library commands have standardized on op code 0x37 for the command
   INITIALIZE ELEMENT STATUS w/RANGE.  Previous implementations by numerous
   vendors had used vendor-unique code 0xE7 - which was implemented by
   NetBackup.  SCSI-3 libraries will return sense data 20/00 when sent an E7.

Workaround:
   To avoid this issue, use only tapes with barcodes that are valid for your
   library vendor.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NetBackup was not properly using the firmware revision to retrieve the
   correct device mapping in the device mappings file.  This has never been a
   problem in the field because VERITAS has never had to use the firmware
   revision in the device mappings file - this is being done for new device
   support.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NetBackup was not properly examining the SCSI library mailslot status bits;
   therefore, during ejects of more than a MAP-full of media, the eject
   operation may abort after the first MAP-full is removed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   bpSALinfo needs to be linked with a more recent set of SAL API libraries to
   be able to interface to ccstor 4.0, and SPC 3.4. Minor changes are needed
   to account for differences in return values for ccstor 4.0.

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   Upon exiting, bpfis closes all file descriptors which includes the file
   descriptor that Hitachi Shadowimage is using. This causes 'bpfis delete' to
   hang for ShadowImage snapshot backups.  

   (NetBackup Clients:  HP Solaris Solaris9 )
--------------------------------------------------------------------------------
Description:
   Previously, cluster_config -a would go into an infinite loop when
   configuring add-on products for monitoring when anything but a lowercase
   "y" was entered at the following prompt:
   
   You have chosen to monitor  <addon>. Is this list correct? [y,n] (Y)
   
   This issue has been resolved and the defaults, "Y" and "y" are all valid
   responses.  

Workaround:
   To avoid this issue, you would have entered a lowercase "y" instead of an
   uppercase "Y", or you could have taken the default.  

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   While trying to reduce the size of a database, the command "bpimage -
   cleanout -client dill-bk" was used.  Using the command in this way caused
   bpimage to core dump.

Workaround:
   To avoid this issue, allow the nightly clean-up process to clean the
   catalog.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   This patch contains the following fixes and additions:
   - Synchronised NetBackup 5.0 MP4 device mappings with 5.1_1.
   - Fixed serialization differences between CSM 3.01 and 4.01 firmware.
   - Added IBM DDS4 autoloader.
   - Added Sony SDX-900V (AIT-4).

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The condition where folders are not included in the backup is only noticed
   during restore operations when attempting to select the folders/files for
   restore. The folder, sub-folder, and/or files are missing from the backup
   selection. This is a randomly occurring condition that does not occur in
   every NetBackup environment. The problem is related to a suspected memory
   corruption and may occur without notice or warning.
   
   It is very important to note that the error may never occur because not all
   Windows systems are susceptible to the memory corruption.  In addition, not
   every system will skip folders and files during the backup.
   
   This issue applies to:
   - MS-Windows-NT and FlashBackup-Windows policies for both full and
     incremental schedules.
   - Any Windows client using VSP or VSS.
   - Windows Open File Backup (WOFB) and Advanced Client backup.
   
   Although VERITAS has been unable to duplicate this issue, the following
   conditions have reported :
   
   a. The policy must contain a file list stating the drives in alphabetical
   drive order.
   
   b. The policy must not use Multi-Streaming.
   
   c. The Client must be configured for individual snapshots.
   
   This issue is not noticed when multi-streaming, if the file list directive
   is ALL_LOCAL_DRIVES, or when using a global snapshot.
   
   If all of the above conditions are met, the backup will include only root
   or top level directories, no sub folders or files.  In severe cases, only
   the drive letter is visible in restore selections.

Workaround:
   Multiple work-arounds are available that reduce but do not guarantee
   complete success if the Client is susceptible to the memory corruption.  
   The following list describes some of the available options.
   
   1. Enable multiple data streams.
   2. Change the client to use a global snapshot.
   3. If using 5.0GA MP1, uninstall MP1 and execute the backup schedule with
      NetBackup 5.0 GA.
   4. Disable WOFB and Advanced Client backup methods.

   All NetBackup Windows Clients
--------------------------------------------------------------------------------
Description:
   When running Windows Open File Backup enabled backups using VSS on
   Windows 2003 clients, performance of the backup is considerably slower
   compared to when running a Windows Open File Backup enabled backup using
   VSP or with Windows Open File Backup disabled. Only backup performance is
   affected and no data loss occurs as a result of using VSS as the Windows
   Open File Backup snapshot provider.

   All NetBackup Windows Clients
--------------------------------------------------------------------------------
Description:
   During multiplexed restores, when multiple tar instances are restoring
   files onto the same directory tree, some of the files may not get restored

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   When selecting a media ID for backup, if the allocation date in the
   media database does not match the assigned date in the media manager
   database, FREEZE the media because there could be database inconsisties.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A large number of down storage units would result in and apparent
   scheduler hang. In reality, the scheduler was timing out to each media
   server in series. This fix recognizes and skips duplicate attempts to a
   down media server..

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   With an AIX master server, FlashBackup on Windows restores were failing
   when a zero length file was encountered.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   It was possible that the bpjobd process may hang under certain conditions.
   It has been demonstrated that this can happen if bpduplicate and bptm
   processes are killed, but it may occur in other situations.  The bpjobd
   process does not die, it just loops consuming CPU time, and nothing
   is written to the debug log.  This prevented the job DB from being updated
   with the status of existing jobs, or new jobs.  bpjobd must be killed and
   restarted by the system administrator to resolve this problem.
   
   This problem is indicated by the following data in the bpjobd debug log,
   referring to non-existent job ID 3:
   
   18:45:00.734 [17433] <2> process_active_job: Frame type is 3 (jobid 3) Sock
   (93)
   18:45:00.734 [17433] <2> process_active_job: Adding provider(1) for job(3)
   socket (93)
   18:45:00.734 [17433] <2> process_active_job: Begin to write (JOBTRYFILE) at
   offset 68 job (3)

Workaround:
   To avoid this issue, kill and restart the hung bpjobd.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The performance of single-file restores for FlashBackup on Windows is below
   expectations.  Increasing the holding buffer size decreases the number of
   call to windows function which in turn decreases the overall restore time.

   All NetBackup Windows Clients
--------------------------------------------------------------------------------
Description:
   For restores of encrypted files with FlashBackup on Windows, the data for
   these files has to be restored sequentially.  The buffering of this data
   was not functioning correctly.

   All NetBackup Windows Clients
--------------------------------------------------------------------------------
Description:
   If AIX Fast Connect is running in a multi-byte environment the LC_MESSAGES
   environment variable is set to C@lft.  This causes English to be used for
   all program output.  If this is encountered, the LANG environment variable
   is tested.  If a failure still occurs, the C locale is used.

Workaround:
   If the LANG environment variable cannot be properly set then the
   /usr/openv/msg/.conf file should be edited and the follow line inserted.
   
      TM 5 C@lft 2 YoUrLoCaLe
   
   where YoUrLoCaLe is the two character local that is used within NetBackup.
   
   For example, "ja" is used for any Japanese locale and "zh" is used for any
   Chinese locale.

   (All NetBackup Servers: RS6000)
--------------------------------------------------------------------------------
Description:
   To assist image database cleanup in 24x7 environments, image database
   cleanup will always run in background.  bpsched will not wait until the
   cleanup is complete before scheduling new jobs.  In addition, bpsched will
   initiate an image database cleanup at periodic intervals, 12 hours by
   default.  The time interval may be changed by using the
   bpconfig -cleanup_int <hours> command.  For example, to change the interval
   to 24 hours, execute the following command:
   
      bpconfig -cleanup_int 24
   
   Use a value of 0 to disable bpsched from starting the periodic image
   database cleanup.  The maximum value is 744 hours.  Bpsched will still
   perform an image database cleanup at the end of a session, however, it will
   be run in background.  On UNIX systems, bpadm may also be used to change
   the interval.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Loaded the correct libxm binaries so that an HP master server will work
   with FlashBackup on Windows.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A synthetic backup time out would occur because of the time it would take
   to read the large extents.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When a user imports and/or restores a NetBackup 3.4 backup image in
   NetBackup 5.0/5.1, files greater than 1GB in size fail to restore and
   produce, "disk full at byte xxxxx", errors in the restore log. Image verify
   also fails with a gigabytes size mismatch error for those files. In
   addition, the Backup, Archive, and Restore GUI shows a discrepancy in
   reporting the size of those files.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   In attempting to filter out file system names that are not appropriate for
   VSM management (for example, /var), VSM-Java filtered out some file system
   names that it should not have filtered (such as, /variety).

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:
   The scheduler debug log was filling with the following message:
   
   "storage unit was not assigned for clientjob <host name> <policy name>
   <stunit name> <copy>"

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The file list for stream 1 is listed twice in the job details when stream 1
   is force-started to take a snapshot. This error is only on multi-stream
   snapshot backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Only fails on a UNIX master. If a user archive fails to remove some files
   (and generates a status 4), then the .f file is removed, and the files, up
   to the failed file, are removed from the client.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The predict algorithm was not sorting correctly. It failed to position
   cumulative incremental backups before the differential incremental backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A multiplexed TIR duplication would fail if the prior backup duplication
   had a TIR size greater than 1 Gigabyte, and would lead to a bptm crash.

Workaround:
   Retry the duplication using non-multiplexed duplication.  It is also
   possible that the duplication will succeed by trying multiplexed
   duplication.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The backup job start time in the activity monitor was the same value for
   each attempt. The start time value should be different on each attempt.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The scheduler would send a message on the wrong queue when it detected an
   error on a client snapshot. This may result in a scheduler crash, if the
   error happened frequently, or if the scheduler stayed running for a
   prolonged period of time.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An intermittent problem with the installer no longer occurs when trying to
   install to remote clients.  The following two files were not being copied
   to %TEMP%\X86, and caused the following error to appear.
   
   Files:  - VERITAS Maintenance Pack.msi
           - Data1.cab
   
   Error message: "The system cannot find the file specified" on Install GUI.

Workaround:
   To avoid this issue, do the following:
   
   1) Start the installer.
   
   2) Select "Select from available computer on the network".  
   
   3) Click Next.
   
   2) Copy VERITAS Maintenance Pack.msi Data1.cab to the %TEMP%\X86 directory.
     (easiest way to find what %TEMP% is, is to open a cmd shell and run echo
     %TEMP%, usually its something like C:\Documents and
     Settings\<UserName>\Local Settings\Temp\X86).
   
   3) Proceed with installation.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The image resume time limit was not always being updated on Checkpoint
   backup jobs. Multiple attempts to resume a job might result in a backup
   that is restarted instead of resumed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The error "license use has been exceeded", status code 159, may have been
   erroneously generated when using a SAN media server license for a media
   server that is part of a cluster.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   BLI failed to mount a clustered storage checkpoint.  The bpbkar log
   displayed the following:
   
   14:56:41.914 [13792] <4> bpbkar set_clone_context: /usr/sbin/mount -F vxfs -
   ockpt=NetBackup_full_test_1096401397 /dev/vx/dsk/cfs_dg/rb_test:NetBackup_fu
   ll_test_1096401397 /rb_test_s/NetBackup_checkpoints/NetBackup_full_test_1096
   401397 >/dev/null 2>/dev/null
   14:56:42.409 [13792] <16> bpbkar set_clone_context: ERR - Cannot mount
   checkpoint NetBackup_full_test_1096401397
   on /rb_test_s/NetBackup_checkpoints/NetBackup_full_test_1096401397.

   (NetBackup Clients:  HP RS6000_433 Solaris )
--------------------------------------------------------------------------------
Description:
   There are two issues that may have caused backups to fail:
   
   1. bpsched was not updating the RESUME_EXPIRATION time on failed backups.
   2. bpdbm did not update the REQUEST_PID on subsequent Q_IMAGE_NEW requests.

Workaround:
   To avoid this issue, make sure any job that has been resumed is finished
   before the cleaning is done.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An initialization failure against nbwin.exe was generated when a
   non-administration account initiated a restore.  A pop-up was generated and
   stated that nbwin.exe failed to initialize 0xc0000022.  A change to the
   permissions was made to msvcr71.dll so that the user "Users" had Read and
   Execute permissions.

Workaround:
   To avoid this issue, manually change the security permissions on
   msvcr71.dll to allow Read and Execute permissions for "Users" user.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The method of configuring inetd has changed in Solaris10. Changed scripts
   that deal with the modification of inetd.conf to use the new method if on a
   Solaris 10 machine (Sparc or X86).

   (All UNIX NetBackup Servers, All NetBackup UNIX Clients)
--------------------------------------------------------------------------------
Description:
   Add support for RedHat 3.0 on the Zseries platform.  This required a new
   client type which is called IBMzSeriesLinux2.4.21.  The number is based on
   the operating system (OS) kernel.

   (All UNIX NetBackup Servers, NetBackup Clients:  new)
--------------------------------------------------------------------------------
Description:
   Added support for RedHat 3.0 on the Zseries platform.  This required a new
   client type which is called IBMzSeriesLinux2.4.21.  The number is based on
   the operating system (OS) kernel.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   This version of the SG driver adds support for Solaris 10.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Description:
   An extra invocation of a function was resetting the delay time on tape UN
   mounts for user backups. Removal of the extra invocation should produce
   more optimal UN mounting of tapes when performing user backups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   This change corrected the scheduler and eased the addition of future
   Microsoft clients.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The scheduler would take a long time scheduling multiple data stream jobs
   if there were many policies that were due to be autodiscovered at once.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If an SSO drive is DOWN on all servers, and its scan host unregisters
   (services are shutdown), vmd/DA will not transfer this drive to another
   scan host.  If the drive is brought UP on a server(s), it will have no scan
   host, and will stay in the NO-SCAN state.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Because of a race condition, ACSLS drives may get downed during mount
   processing.

Additional Notes:
   The signature of this race condition is that the event viewer (or syslog)
   will note errors in semaphore operations from ACSD, and then the drive will
   be DOWNed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   For Disk Staging, NetBackup could not clean all images when the total number
   of images were more than 16384.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Re-naming a schedule, and then attempting an immediate backup using the
   newly named schedule caused a core dump in the scheduler.

Workaround:
   To avoid this issue, do not add, remove, or modify policy or schedule
   information and then attempt an immediate backup while scheduler is running.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Tapes would not unmount in a reasonable time frame.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   In a rare circumstance, the last block of a multiplexed, duplicated copy of
   a TIR backup may be corrupt. A verification of the copy would not be
   successful. The original backup is not affected. This does not affect
   non-multiplexed duplications of a multiplexed TIR backup.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected a potential problem where Windows Open File Backup (WOFB),
   multi-stream, non-checkpoint, non-control stream gets a 134 status, and
   stays queued (never goes active).

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Maintenance Pack 3 for the NetBackup 5.0 release introduced a performance
   issue with NDMP DAR restores for catalogs with a large number of files.
   This fix corrects the performance problem.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected a case where incremental backups were backing up complete data
   sets (full backup). This happened only for multi-stream, Windows Open File
   Backup (WOFB), or advanced client where a previous non-control stream
   backup failed.

   (All NetBackup Servers)


=========
NB_50_3_M
=========
Description:
   bptm fails with a segmentation violation when the media Disk Staging
   Storage Unit sends a duplication job to the master.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected an end-case problem in the MPX read processing
   (restores/duplications) that would try to unmount a media that is already
   in use by another process.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added SDLT-600 drive type and SDLT_2 media type support to ACS robotics.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Media position error while attempting to restore from a copy of a backup
   created by Inline Tape Copy.  The bptm log will show an entry similar to
   the following (note the file number of 0):
   
   13:12:27.348 [22187] <2> mpx_read_init: media id A00001, copy 1,
   fragment 1 (839680 Kbytes) filenum 0 being considered for restore
   
   The problem occurs when performing multiplexed backups using Inline Tape
   Copy, and a new backup is started while waiting for a tape mount due to
   hitting end-of-media.  

Workaround:
   The restore may be done from the other copy.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A CheckPoint Restart backup may hang after resuming an incomplete backup.  
   This would happen if the backup error occurred after backing up more than
   four gigabytes of data.
   
   The bptm log will show many entries similar to the following:
   
   09:40:43.700 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508
   09:40:43.730 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508

   09:40:43.760 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508
   
   ...where the calculated blocks value and the blocks in cpr value do not
   change.

Workaround:
   Disable the CheckPoint Restart feature in the policy, or create a maximum
   fragment size in the storage unit that is less than four gigabytes.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The timeout for an ACS query is configurable using the ACS_QUERY_TIMEOUT
   entry in vm.conf, (value in minutes).
   
   Example:
   ACS_QUERY_TIMEOUT = 30
   
   This sets the timeout to 30 minutes.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   VOPIE Enhanced Authentication may not work when host names are greater than
   32 characters.  You may receive Authentication Failed (error code 160)
   messages when connecting to systems with host names or client names greater
   than 32 characters.

   All NetBackup Windows Clients All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Disk Staging would not correctly relocate backup images from clients when
   a client name contained an "_" character.

Workaround:
   To avoid this issue, remove the "_" from the client computer names.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The following two issues with checkpoint restart for backups have been
   resolved.
   
   1. The import of a successful checkpoint restart backup that is incomplete
   or suspended prior to the first checkpoint may fail.  Then, if it is
   resumed, it will complete successfully.  The phase 1 import log would show
   one or more of the following entries:
   
   13:45:45 INF - Found existing image information for client test, policy
   itest, and schedule Full, done on 04/13/2004 13:37:56.
   13:45:47 WRN - Cannot import client test, policy itest, and schedule Full,
   copy 1 fragment 1 already exists. Skipping this image.
   
   The phase 1 import will then succeed.  However, the phase 2 import will
   fail with the following error:
   
   13:56:19 INF - Import of policy test, schedule Full (test_1081881476)
   failed, tar had an unexpected error.
   
   
   2. For a checkpoint restart backup, in the rare case where bptm detects
   that a position check problem occurred following a checkpoint due to a
   misconfigured drive or a rewind from an external source, and the backup is
   later resumed, the information on the tape prior to the checkpoint may be
   invalid.
   
   The bptm log would indicate the position check problem with one of the
   following logs following a checkpoint:
   
   08:39:57.969 [4393] <16> write_data: FREEZING media id 00011, too many
   data
   blocks written, check tape/driver block size configuration
   
   OR
   
   log.041204:14:39:12.373 [6416] <16> write_data: FREEZING media id 00005,
   External event caused rewind during write, all data on media is lost
   
   The problem would occur if the same backup were resumed and completed with
   a successful status.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Exchange restores from multiple images fail with a communications failure
   error. The log will have a message similar to the following:
   
   17:58:45 (158515.002) ERR - unable to create object for restore:
   Microsoft Information Store:\SG01102\Log files_1074011439 (BEDS
   0xFFFFFE30: A communications failure has occurred.

Workaround:
   To avoid this issue, create a file named,
   /usr/openv/netbackup/NON_MPX_RESTORE to force sequential restores from
   multiple images.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A possible buffer over-run corrupting eject information in memory, was
   causing ACS ejects to fail because the CAP was reported as not existing,
   when it actually did exist.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A NetBackup restore that has a path specified to be excluded from restore,
   but that path does not even exist in the backup image, was completing with a
   partial success status of 1. The resolution of this DDTS would complete
   such a backup with a complete success status of 0.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When using NetBackup Access Control hostnames, a specific size may cause an
   erroneous Authorization denial.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Some multiplex groups are lost on Windows 2000 systems only. This results
   in the inability to cancel/suspend jobs or add new jobs to the multiplexed
   group.  In addition, on all platforms, the overhead on starting an
   immediate, user, or archive backup is excessive.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A potential buffer over run in scheduler has been resolved.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   For FlashBackup on Windows, restores of large files were failing.  This
   happened if the extents composing the files described more than 2GB of
   data.  In a situation like this, an extent larger that 2GB was broken down
   into smaller units but not all of the fields associated with that extent
   was propagated correctly.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bogus links are created when loading Linux and SCO clients if the INTEL
   directory exists.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Description:
   When the NBAR starting and stopping scripts moved to netbackup.sh,
   bp.kill_all was never updated to stop the daemons the same way.  It is
   still looking for the scripts in "/etc", "/sbin", etc. where they used to
   reside.  The bp.kill_all command still kills the scripts via the
   second_chance, however it asks a question before proceeding with the kill
   operation.
   

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   When running some of the client commands on an Irix 6.5.22m system, they
   may core dump.  This is a known problem for SGI and to workaround the
   client binaries must be linked with -lC.

   (NetBackup Clients:  SGI65 )
--------------------------------------------------------------------------------
Description:
   On a system running Windows 2000 (SP4), you may not be able to browse or
   restore the second volume if 'SET HIST=N' is listed in the middle of the
   policy's file list.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Queries for images having XBSA extended data fail in bprd when connecting
   to dbm with the socket being overwritten.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The restore of a backup created by NetBackup 4.5, or earlier NetBackup
   releases may fail on NetBackup 5.0.  This defect applies to backups of a
   raw partition and backups of a file whose size is 1GB or greater.
   
   The failed restore may return an exit status of 0, and it may otherwise
   appear to have succeeded.  However, the restored file will not contain all
   the data of the original file.  Alternatively, the restore may exit with a
   status of 3 or 5.  The tar debug log may contain a message of "invalid tar
   header" for the failing file.
   
   In addition, attempts to verify the backup image may fail on NetBackup 5.0.

Workaround:
   TO avoid this issue, do the following:
   1. Expire the backup image.
   2. Import the backup image.
   3. Run the restore.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Catalog archiving does not properly handle images written by
   NetBackup 3.1.1.   Using the archive tools with images that do not have a
   CATARC id can result in a catalog file being backed up and removed with out
   the CATARC id being written.

Workaround:
   Adding the line "CATARC 0" the the catalog image file before including it
   in a bpcatarc call will avoid this problem.

Additional Notes:
   If the images have already been archived without the CATARC value being
   written, you can get the files back by restoring the files from the backup
   that resulted from the initial bpcatarc call.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The bpexpdate binary would not notify the user that they were unable to use
   bpexpdate due to access control limiting their access. Although the binary
   would return the appropriate error level (such as, 117, 116, or 118), the
   corresponding message contained unreadable characters.

Workaround:
   check the error level returned by bpexpdate manually. (e.g. echo
   %ERRORLEVEL% or echo $? )

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NBU_Operators did not have the backup permission on Policies, yet they were
   able to perform manual backups. The NBU_Operator group should be able to do
   manual backups, but it should always require the Backup permission on
   Policies.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An issue existed in the command bpcatlist | bpcatarc | bpcatrm.
   
   After successful catalog backup, none of the catalogs are deleted.  If you
   do a bpcatlist, you will notice that there is a "0" in the Catarcid value.
   In the policy_time_FULL image file, CATARC is 0.  It appears that this is
   not being updated.  You are still able to browse files for restores.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   In cases where a master server was set to require NetBackup Access Control
   (NBAC) and a client was set to automatic without a credential, the HP-UX
   platform would not return the same error code as other platforms. In such
   cases, error codes 192 and 193 would be documented in the debug log, but
   would be exposed back to the user as error code 23.

Workaround:
   To avoid this issue, set bp.conf entry USE_VXSS = PROHIBITED or
   USE_VXSS = REQUIRED on the client system.

   (NetBackup Clients:  HP )
--------------------------------------------------------------------------------
Description:
   xbp was inserting and deleting items from lists one item at a time. This
   was done because of deficiencies in some X-Servers' support of 1.1 Motif.
   This is not an efficient way to handle lists with a large number of items.
   
   With X-Servers that support Motif 1.2 this is not an issue. The code was
   changed to delete all items at once and to insert multiple items at the
   same time.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   The number of in-use drives is not computed correctly when a disk staging
   storage unit is flushed (duplicated) to tape. This results in additional
   backup jobs being blocked until the duplication is completed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Boolean file keys in Device Mappings are not properly parsed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   There was a possibility that a core dump could occur in the scheduler if
   the storage unit host name cannot be mapped to a valid media server name.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   During an Exchange backup of the Information Store, the Bandwidth
   Limitation feature of NetBackup is not supported. Backups of Exchange
   proceed at the fastest pace possible for the network.  
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When running multiple, multiplexed and multi-streamed backups using
   NetBackup 5.0 MP1 or NetBackup 5.0 MP2, some of the backup jobs could end
   with status 69: invalid filelist specification. The file list will contain
   a NEW_STREAM directive which is causing the status 69 errors.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Assigned a new message ID.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   bpdbm's host list is not consistent with that of bprd. The current
   impact of this issue is  that "FORCE_RESTORE_MEDIA_SERVER does not work for
   TIR restores."
   
   It appears that FORCE_RESTORE_MEDIA_SERVER does not replace the server
   entry for TIR fragment. Thus, when TIR information needs to be extracted
   from tape, bpdbm tries to initiate "bptm -restoretir" on the original
   server.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added support for the following drives:
   Certance LTO2 drive added, HP SDLT600, Sony SDX-700V, HP C7438A DAT 72,
   
   Added support for the following libraries:
   Qualstar RLS Series Libraries, Quantum DX100, HP MSL6000, HP MSL5000,
   QUANTUM ValueLoader
   
   In addition, several changes have been added to address serialization,
   library inventory, and extend/retract issues.  The following list
   identifies the library and the issue that has been corrected.
   
   ADIC Scalar i2000 - extend retract issue
   Compaq MSL5000 - barcode inventory issue.
   OVERLAND Libraries - barcode inventory issue.
   Fujitsu LT160 Eternus  - empty-map and serialization issue
   
   Added Support for the following drives:
   Certance LTO2 drive added, HP SDLT600, Sony SDX-700V, HP C7438A DAT 72,
   
   Added Support for the following libraries:
   Qualstar RLS Series Libraries, Quantum DX100, HP MSL6000, HP MSL5000,
   QUANTUM ValueLoader
   
   In addition, several changes have been added to address serialization,
   library inventory, and extend/retract issues.  The following list
   identifies the library and the issue that has been corrected.
   
   ADIC Scalar i2000 - extend retract issue
   Compaq MSL5000 - barcode inventory issue.
   OVERLAND Libraries - barcode inventory issue.
   Fujitsu LT160 Eternus  - empty-map and serialization issue

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When a backup job using Checkpoint Restart with a file list of
   ALL_LOCAL_FILES either fails or is suspended before the first checkpoint is
   taken, then the resumed backup job will skip the backup of the first local
   file system it would have backed up, and exit with status code 0.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Restoring symbolic links owned by non-root could show an error even though
   the symbolic link was successfully restored (giving Partial Success for the
   restore job).

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   During disk staging duplication, bpduplicate prints the following message
   although no error has occurred.
   
   11/22/2003 11:25:35 AM - Error bpduplicate(pid=1636) from host dcdell1,
   COMPLETE_MOUNT QUANTUMDLT70000 0

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Backups hang when backing up FIFO files on VxFS 4.0 file systems on Solaris
   6, 7, and 8.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Correct the case where the 's t' command within tldtest returns incorrect
   data when the number of transport is 0.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Provide the ability for the user to control the format of the drive names
   that are created by the Device Discovery Wizard.

Additional Notes:
   The following variables could be used in generating a drive name            
   
          o HN - hostname of the server where the drive is first discovered
          o RN - robot number
          o RT - robot type
          o POS - Drive position, determined by robot type.  One of the
   following:
                 - robot drive number
         - vendor drive id
         - ACS.LSM.PANEL.DRIVE
          o SN - serial number of the drive
          o DT - drive type  (i.e. hcart, 8mm, dlt2)
          o VEND - vendor ID of the drive obtained from the inquiry string of
         the drive
          o PROD - product ID of the drive obtained from the inquiry string of
          the drive    
          o literal strings proceeded by a pound sign(#) containing the
    allowable characters in a drive name.
          o IDX - ensure an index is appended. Otherwise, an index will be
                 automatically be appended if needed to make a drive name
         unique.
   
    Note: Any whitespace in the above entries will be collapsed.
   
    These entries will allow a tiered approach.  If the DRIVE_NAME_SEED
    entry appears on a non-global-database host, that format will be
    used.  Otherwise, the vm.conf file on the global database host will
    be checked for a DRIVE_NAME_SEED entry.  If one exists, it will be
    used.  If none exists in either location, the default naming scheme
    will be used. For tpconfig this is "Drive" followed by a number to
    make it unique.  For tpautoconf, this is VendorID, followed by
    ProductID, followed by a number to make it unique.  
   
   EXAMPLES:
   1) To generate the default names used by discovery provide the entry:
               
    DRIVE_NAME_SEED = VEND:PROD:IDX
   
      This will produce drive names of the form
       
    QUANTUMDLT70000
   
   2) To generate the names names that contain the drive's product ID and
   robot residence information, fields separated by minuses:
   
           DRIVE_NAME_SEED = PROD:#-:RT:#-:RN:#-:POS
   
      This will produce drive names of the form
       
    DLT70000-TLD-0-1
   
   Issues and Limitations
   
   Some of the below are just general drive naming issues, while others
   relate specifically to this feature.
   
   1. In a shared storage (SSO) environment drives will be named based on
      the seed of the first host that discovers them.  Subsequent hosts
      wlll retain that drive name regardless of the value of their local
      DRIVE_NAME_SEED vmn.conf option.  
   
   2. The Hostname (HN) should be used with caution in SSO
      configurations.  Drives shared amongst hosts must have the same
      name.  See 1 above.
   
   3. Drive names remain limited to 48 characters.
   
   4. Valid characters are upper and lower case alphanumerics, plus,
      minus, underscore, and period.
   
   5. Drive names cannot start with a minus (-) character.
   
   6. If robot residence information is used to generate the drive name, the
   host that has the robot attached should be discovered first.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Compressing the 0-byte catalog files on a Tru64 system creates 0-byte, .Z
   files.  On other operating systems the file is 3 bytes in size.  When
   running uncompress, it fails with an error, that it is not in a compressed
   format.

Workaround:
   To avoid this issue, ignore the error messages in the log, or implement a
   custom compress script.

   (All NetBackup Servers: Alpha_5)
--------------------------------------------------------------------------------
Description:
   A problem existed when uninstalling NDMP that left a namespace.chksum entry
   in the *NBU_RSP on clusters. Because the uninstall removes this file on the
   shared disk, the monitor will report a fault.  

Workaround:
   To avoid this issue, manually remove the following line from the RSP file
   found in /usr/openv/netbackup/bin/cluster/<platform>/*_NBU_RSP.
   
   LINK=volmgr/database/.namespace.chksum

   (All NetBackup Servers: Alpha_5 HP Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   Localized characters were being turned into unreadable gibberish by a
   third party charting API.  The File System Analyzer (FSA) would then
   display this translated text as gibberish.  To avoid displaying the
   unreadable characters, the localized characters are no longer passed to the
   third party charting API.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected a typographical error in bpdbm by changing "sysnthetics" to
   "synthetics".

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Flashbackup single file restores did not start. In addition, bprd was slow
   sorting the file history.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Restart of frozen image backup (non-WOFB) can fail. The scheduler
   erroneously clears the snapshot information prior to scheduling the backup.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Performing backups using an include list with non-fully-qualified path
   names could cause significantly slower backup performance.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   The following changes have been made:
   - Added error handling for blank barcodes and missing media magazines
   on DELL PV-132T.  
   - Changed the media type for IBM 3592J to hcart2.
   - Added/Changed error handling for blank barcodes and missing media
   magazines on IBM 3582, COMPAQ MSL5000, COMPAQ SSL2000, IBM 3607
   to fix inventory issues.
   - Add Sony SDX-400V, SDX-500V drives.
   - Add Qualstar RLS-4124, RLS-4445, RLS-4470, RLS-5244, RLS-6227, RLS-8116,
   and RLS-8236.
   - Fixes inventory issue for some STK libraries when MAP is empty and
   inventory is done with empty MAP selected.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When using "bpexpdate -recalculate" or "bpexpdate -backupid" to extend the
   expiration date of one or more backups that have multiple copies, the media
   expiration may not be correct for copies on removeable media if the first
   copy is on disk.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Backups with exclude and include lists can incorrectly back up excluded
   directories (although not any files within those directories).

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Using the options that allow a UNIX file system to be restored without
   crossing mount-points fails by restoring all files under the path
   instead of just the single filesystem.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An occasional failure occurred when duplicating True Image Restore (TIR)
   backups.  The failure may have occurred if there was a delay in mounting
   the write media.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   With STK's new verison of SSI, certain environment variables are required
   to be set when the vm.conf entry ACS_TCP_RPCSERVICE is set.  The variables
   are set to default behavior unless they are set in the vm.conf.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected a syntax error in bpbbackupdb.c.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A regression was introduced in NetBackup 5.0GA (and continued in
   NetBackup 5.0MP1) whereby the bpcatlist command will not work properly with
   the "bpcatlist -client <client-name>" syntax.
   
   The following commands, "bpcatlist -client all" or "bpcatlist -client
   <valid-client-name>" will return a status 227, (No entity was found).

Workaround:
   The use of this command cannot select a specific client to archive the
   information about, but may archive on other attributes.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Resume of a single-stream suspended job would backup the complete policy
   file list instead of only the stream's portion of the file list.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bptm or bpdm may crash after bpsched restarts following a bpsched crash
   and
   subsequent resuming of checkpointed backup jobs on media servers.  This
   change will cancel existing duplicate jobs when this situation occurs, or
   prevent the crash from occurring.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When scheduler verbose is set greater than zero, there is the potential for
   Jobs to exit with a status 50. This is more likely on systems experiencing
   connection issues. The situation occurs when the scheduler prints a message
   and loses the errno value. Armed with the incorrect errno value, scheduler
   exits (status 50) the current running jobs.

Workaround:
   To avoid this issue, set scheduler verbose to zero.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When the primary copy of the data image is expired, synthetic backup fails
   even though there are valid code exists.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Following bpsynth messages were not getting localized.
   
   "unable to receive the response to the image query, error=%s(%d)"
   "caught synth EXCEPTION, error=%s(%d)"
   "preparation for generating synthetic image failed,
   error = %s(%d)"
   "no images were found to synthesize"
   "end bpsynth with status ="  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NetBackup patch NB_50_2_M_268356.winnt.intel.exe can provide miss-matched
   versions of the netbackup.dll after applying the MP2 patch.  NetBackup
   processes that access the netbackup.dll while the patch is being installed
   will cause the installer to not update the correct version of netbackup.dll
   in other locations (such as, Volmgr/bin, NetBackup/bin/goodies,
   NetBackup/bin/admincmd).
   
   Microsoft Event log error message looks like the following:
   
   Event id 7024 - "The NetBackup Device Manager service terminated with
   service-specific error 1. " Event ID 4234 - "unable to load bpnp.dll. The
   specified module cannot be found."

Workaround:
   To avoid this issue, manually copy the correct version of NetBackup.dll from
   VERITAS\NetBackup\Bin to the other locations (if they exist):
   
   Volmgr/bin,
   NetBackup/bin/goodies,
   NetBackup/bin/admincmd  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Systems with multiple tld or multiple tl8 robots may see strange behavior
   across robot numbers on remote (not control hosts) media servers when the
   daemon is bounced.  In the worst case, this can include the drives going
   into "AVR" control mode on the noncontrol host media servers.  Generally
   this is started by a catastrophic failure of one of the robots.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected a restore problem from the disk where a status 92 would occur if:
   - The image contained multiple disk fragments and first file of the restore
     was not in the first fragment.
   - The size of the fragments, prior to the one containing the first file,
     were not a multiple of 1K.
   
   The above conditions can occur during a MPX multiple copy backup, one
   or more copies to disk, and one or more copies to tape.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected potential scheduler problems with memory leaks and non-
   initialized malloc space.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Adjusted a loop used during the deletion of a multiplexed group. This loop
   was not functioning correctly on the Windows 2000 platform and resulted in
   the removal of additional multiplexed groups.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bprd hits a segmentation fault and core dumps. The issue appears to
   occur when the log file path for the bprestore log exceeds 58 characters.
   This only occurs when the media server involved is not the master server,
   and when FQDN names are used.
   
   To resolve the issue, increase the command buffer size to BUFFLEN (4096),
   and implement bounds checking on all string buffers, where practicable,
   with the use of V_strlcpy, V_strlcat, and snprintf().  

Additional Notes:
   While libVcom.so contains no changes, it is required to resolve new entry
   points introduced in libsfr.so.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added Solaris 10 to drop-down menu.

Additional Notes:
   For "Solaris Solaris10" to show when configuring clients into a backup
   policy, the user must execute the following command:
   
   /usr/openv/netbackup/bin/goodies/new_clients

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Resume of a checkpoint backup could select a different media server if
   failure on initial backup attempt.
   
   Failure scenario:
     1) Backup reached a checkpoint.
     2) Then failed (this case was a 13 "read failed").
     3) Scheduler attempts to programmatically re-submit.
     4) But the drives were re-set & downed.
     5) So, the job exited with a 213 "no storage units available".
   
   The original storage unit information was "lost" so the job could
   potentially resume on a different media server. If resumed in this way,
   the backed up data would reside on two different media servers and any
   subsequent restore would fail.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Changes needed for Solaris10 support.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   After an NDMP import, the file count drops by a very Large number.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   After NDMP import the file count drops by a very Large number.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The potential for missed files can occur if all the following conditions
   exist:
    1) Advanced client backup.
    2) Incremental backup.
    3) Multi-stream backup.
    4) Backup stream is non-stream 1.
   
   Files may be missed if they are modified after the snapshot is taken but
   before the (non-stream 1) image is created.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The resume of a checkpoint backup could select a different media server if
   a failure occurred on initial backup attempt.
   
   The following list summarizes this type of failure scenario:
     1) Backup reached a checkpoint.
     2) Then failed (this case was a 13 "read failed").
     3) Scheduler attempts to programmatically re-submit.
     4) But the drives were re-set & downed.
     5) So, the job exited with a 213 "no storage units available".
   
   The original storage unit information was "lost" making it possible for the
   job to resume on a different media server. If resumed in this way, the
   backed up data would reside on two different media servers and any
   subsequent restore would fail.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   There is a potential for missed files if all the following conditions
   exist:
    1) Advanced client backup.
    2) Incremental backup.
    3) Multi-stream backup.
    4) Backup stream is not stream 1.
   
   Files modified after the snapshot is taken but before the (non-stream 1)
   image is created may be missed.

   (All NetBackup Servers)
================================================================================


=========
NB_50_2_M
=========
Description:
   When using NetBackup Access Control hostnames, a specific size may cause an
   erroneous Authorization denial.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The following two issues with checkpoint restart for backups have been
   resolved:
   
   1) An import may fail if a successful checkpoint restart backup that is
   incomplete or is suspended prior to the first checkpoint and then is
   resumed and completes successfully.  The phase 1 import log would show one
   or more of the following entries:
   
   13:45:45 INF - Found existing image information for client test, policy
   itest, and schedule Full, done on 04/13/2004 13:37:56.
   13:45:47 WRN - Cannot import client test, policy itest, and schedule Full,
   copy 1 fragment 1 already exists. Skipping this image.
   
   The phase 1 import will then succeed.  However, the phase 2 import will
   fail with the following error:
   
   13:56:19 INF - Import of policy test, schedule Full (test_1081881476)
   failed, tar had an unexpected error.
   
   
   2). For a checkpoint restart backup, in the rare case where bptm detects
   that a position check problem occurred following a checkpoint because of a
   misconfigured drive or a rewind from an external source, and the backup is
   later resumed, the information on the tape prior to the checkpoint may be
   invalid.
   
   The bptm log would indicate the position check problem with one of the
   following logs after a checkpoint:
   
   08:39:57.969 [4393] <16> write_data: FREEZING media id 00011, too many
   data
   blocks written, check tape/driver block size configuration
   
   OR
   
   log.041204:14:39:12.373 [6416] <16> write_data: FREEZING media id 00005,
   External event caused rewind during write, all data on media is lost
   
   The problem would occur if the same backup were resumed and completed with
   a successful status.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   FlashBackup restores of large files were failing on Windows platforms.
   This happened if the extents composing the files described more than 2GB
   of data.  In this case, the extents larger that 2GB were broken down into
   smaller units but not all of the fields associated with that extent were
   propagated correctly.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A Checkpoint Restart backup may hang after resuming an incomplete backup.  
   This would happen if the backup error occurred after backing up greater
   than four gigabytes of data.
   
   The bptm log will show many entries, similar to the following example,
   where the calculated blocks value and the blocks in cpr value do not
   change.
   
   09:40:43.700 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508
   09:40:43.730 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508
   09:40:43.760 [2140.3540] <2> write_data: Received checkpoint for backup id
   crowvalley_1079629802, calculated blocks: 13147044 blocks in cpr: 13081508

Workaround:
   Disable the Checkpoint Restart feature in the policy, or create a maximum
   fragment size in the storage unit that is less than four gigabytes.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Media position error may occur while attempting to restore from a copy of a
   backup created by Inline Tape Copy.  The bptm log shows an entry similar to
   the following (note the file number of 0):
   
   13:12:27.348 [22187] <2> mpx_read_init: media id A00001, copy 1,
   fragment 1 (839680 Kbytes) filenum 0 being considered for restore
   
   The problem occurs when performing multiplexed backups using Inline Tape
   Copy, and a new backup is started while waiting for a tape mount due to
   hitting end-of-media.  

Workaround:
   To avoid this issue, you can do the restore from the other copy.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Synthetic backup did not recognize or work compressed images.  As a result,
   a synthetic backup would ignore compressed images.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Cannot restore from a CheckPoint Restart job that was restarted.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When a backup job using Checkpoint Restart with a file list of
   ALL_LOCAL_FILES either fails or is suspended before the first checkpoint is
   taken, then the resumed backup job will skip the backup of the first local
   file system it would have backed up, and exit with status code 0.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   An interrupted Checkpoint Restart backup would time out after the backup
   was resumed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Queries for images that have XBSA extended data fail in bprd when trying to
   connect to dbm while the socket is being overwritten.  This issue has been
   address by increasing the size of the data buffer by 2048 bytes per file to
   account for the XBSA extended data.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   bptm fails with a segmentation violation when the media Disk Staging
   Storage Unit sends a duplication job to the master server.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The restore of a backup created by NetBackup 4.5, or earlier
   NetBackup releases may fail on NetBackup 5.0.  This defect
   applies to backups of a raw partition and backups of a
   file whose size is 1GB or greater.
   
   The failed restore may return an exit status of 0, and it
   may otherwise appear to have succeeded.  However, the restored
   file will not contain all the data of the original file.
   Alternatively, the restore may exit with a status of 3 or 5.
   The tar debug log may contain a message of "invalid tar header"
   for the failing file.
   
   In addition, attempts to verify the backup image may fail on
   NetBackup 5.0.

Workaround:
   1. Expire the backup image.
   2. Import the backup image.
   3. Run the restore.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Catalog archiving does not properly handle images written by
   NetBackup 3.1.1.  Using the archive tools with images that do not have a
   CATARC ID can result in a catalog file being backed up and removed without
   the CATARC ID being written.

Workaround:
   To avoid this issue, add the line "CATARC 0" to the catalog image file
   before including it in a bpcatarc call.  

Additional Notes:
   If the images have already been archived without the CATARC value being
   written, you can recover the files by restoring the them from the backup
   that resulted from the initial bpcatarc call.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If a policy with checkpoints enabled uses a storage unit with a maximum
   fragment size, and the maximum fragment size is reached prior to the first
   checkpoint, the job may not be resumeable.  When the job is resumed, bptm
   fails with a 223 error.

   (All NetBackup Servers)
================================================================================


=========
NB_50_1_M
=========
Description:
   Check to make sure a groups entry exists in the xinetd
   files for the Macintosh 10.1 client.  If one does not exist,
   remove the existing xinetd file and create a new one.

   (NetBackup Clients:  MacOSX10.2 )
--------------------------------------------------------------------------------
Description:
   When starting a backup using the -i (immediate) and -L (logging) options,
   the job does not appear in the bpjobd monitor (Activity Monitor).  
   
   The ability to use backup progress logging on clients when using the
   bpbackup -i command, which is currently disallowed by bprd and bpbackup,
   has been added.  In addition, bpbkar is successfully receiving the
   -L option, indicating that bpsched is passing the log file successfully.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Add an exit 0 after checking for NBAR processes.
   If NBAR is present but not up, the grep causes bpps to
   end with a non-zero exit status.  GDM, at a minimum,
   checks bpps' exit status.

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   bplist processes paths and interprets '..' and '.' in paths.  This does not
   allow the use of a path such as 'x/./y', which NetWare uses.  A change
   was made to allow '/\./' to be interpreted as '/./'.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   The scheduler can get socket read failures if the shell script writes
   error messages to stderr. The bpsched log will have messages like the
   following:
   
   12:26:41.325 [4427] <2> get_string: buffer space for 5121 bytes, but
   incoming data needs 1668248940 bytes
   12:26:41.325 [4427] <2> get_string: (7) inadequate buffer space for data
   12:26:41.325 [4427] <16> readstring: get_string failed: Error 0 (0),
   incoming data too large for buffer (-1)
   12:26:41.325 [4427] <8> read_bpbrm_stderr: readstring failed: socket read
   failed (23)

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   After using the job monitor to terminate a backup, the bpbkar process
   on the client can sometimes hang in exit processing looking for a status
   message that was already sent.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Imports or verifies with detailed logging can take much longer than
   without detailed logging.

Workaround:
   Don't request detailed logging.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Restores of remote clients initiated from a NetBackup server with detailed
   logging can take much longer than restores without detailed logging.

Workaround:
   Don't request detailed logs or initiate the restore from the remote
   client.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   bptm core dumped when the TPC buffer size was set to 256K, and the backup
   included a directory with many files of less than 512 bytes. The TPC
   mover was a Pathlight, which advertises a maximum number of segments per
   command of 256, but will support over 4096.

Workaround:
   To eliminate this failure, reduce the TPC buffer size to 128K or less.

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   Bad characters may appear in the Device Identifier field of the "scan"
   utility output for some devices.  The "scan" utility now displays the
   correct Device Identifier information.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An error message that indicates a failure to obtain device target
   information from the 3pc.conf file is being overwritten by a "device not
   found" error message, thus making it impossible to detect certain
   misconfiguration errors.

   (All NetBackup Servers: HP RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   The parsing of bp.conf in some of the packaging and install
   scripts, when looking for SERVER entries, was not
   restrictive enough and extra entries were being pulled in.
   The parsing has been corrected.

   (All UNIX NetBackup Servers and All NetBackup UNIX Clients)
--------------------------------------------------------------------------------
Description:
   The bpsched process core dumped and left a child process still running. The
   remove_from_mpxgrp_activelist() function attempts to free pool information,
   but the pool pointers are really residue.  Changed add_to_active_job_lists
   () function to initialize the malloc'd structure.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   FlashBackup does not restore VxFS meta data correctly.  For example, if a
   file contains VxFS attributes (such as reserve or fixed extent size), the
   file restore will fail resulting in a zero length file.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added functionality to check for and handle errors when converting from
   char to wchar_t.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   A problem existed with Q_CLIENT_FIM_LIST in the header file. To resolve the
   issue, the Q_CLIENT_FIM_LIST has been reassigned.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A minor change was made to a while loop condition in the SGI sparse file
   processing code to eliminate a possible infinite loop.

   (NetBackup Clients:  SGI65 )
--------------------------------------------------------------------------------
Description:
   bpdgclone failed when a volume has dco log.
   
   online_util log:
   09:33:22.155 [27646] <2> onlfi_vfms_logf: INF -     make_new_vol:
   executing /usr/sbin/vxmake -g clone_hdsdg_clone < /var/tmp/DGCLBAAquaGd2
   09:33:22.156 [27646] <2> onlfi_vfms_logf: INF -     vxvm:vxmake: ERROR:
   Dco my_test_dco not in configuration
   09:33:22.156 [27646] <2> onlfi_vfms_logf: INF -     vxvm:vxmake: ERROR:
   getting associations of :
   09:33:22.156 [27646] <2> onlfi_vfms_logf: INF -     Bad record name
   09:33:22.156 [27646] <2> onlfi_vfms_logf: INF -     could not vxmake
   from input file /var/tmp/DGCLBAAquaGd2
   

   (NetBackup Clients:  Solaris )
--------------------------------------------------------------------------------
Description:
   Problems with barcode errors in exabyte libraries have been fixed.  In
   addition, fixes for cleaning tape recognition in Compaq/DEC rebadges of
   Overland libraries and Sony CSM serialization are also included.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   In NetBackup 5.0, cluster_config would not start the daemons on a NetBackup
   media server or after a master server upgrade.  This issue has been
   corrected so that cluster_config does start the daemons.

Workaround:
   The user must start the cluster services manually using the appropriate
   command for their system. If /usr/openv/netbackup/bin/cluster/.do_templates
   is present, the user must run this script to configure the NetBackup
   templates and then delete it on all nodes of the cluster.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Changes have been made to use version 3.23.58 of MySQL for nbdbd.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The VCS monitor script would change the value of .NBU_STATE if it detected
   processes were offline. This resulted in cluster_active occasionally giving
   an incorrect status and causing failures during upgrades or reinstalls. To
   resolve this issue, changes were made to the monitor so the .NBU_STATE is
   not changed. The NBU_STATE is only changed during an online, offline, and
   clean now.

   (All NetBackup Servers: HP Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   Changes were made to correct problems when using the NetBackup TruCluster
   agent with 5.1B pk2 and 5.1 pk5. These patches from Hewlett Packard put
   brackets around process names and break the caa_netbackup monitor script.  

   (All NetBackup Servers: Alpha_5)
--------------------------------------------------------------------------------
Description:
   NetBackup 5.0GA incorrectly allows the user to select the 'Use data mover'
   attribute for Microsoft Windows NT and FlashBackup-Windows policy types for
   Advanced Client.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NBAC has required use of FQDN host names in the past. Now host names are
   accepted as short or long host names and are converted to long host names
   internally as needed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If an inline tape copy job starts with multiplexing set to 1, another job
   with multiplexing set greater that one will attempt to multiplex with the
   first job.  The multiplex setting of 1 on the first job caused the job to
   run without multiplexing enabled.  The second job will fail because it
   cannot multiplex with the first job.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If a restore fails because of a media mount timeout, bprd may core
   dump when it finishes processing the restore.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Because of the complex nature of timezone and daylight savings time
   interaction, the timezone correction factor for daylight savings time was
   using the wrong value and caused an incorrect check to see if daylight
   savings time is currently in effect.  Changes to the timezone correction
   macro were made to resolve this issue.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The log analysis utilities (backuptrace, backupdbtrace, importtrace,
   restoretrace, duplicatetrace, verifytrace) could not be executed by a non-
   root administartor in an NBAC configuration with NetBackup 5.0 GA. The
   problem was caused by the install permissions on these scripts.
   
   This patch will install these scripts with permissions that will allow them
   to be executed by a non-root administrator. The non-root administrator must
   be defined under NBAC.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A 134 status error would be displayed as an attempt or retry in the
   Activity Monitor even though they are not counted as tries.  A change has
   been made to the scheduler to not send an "end_try" to bpjobd when a
   134 status error is received.  In addition, a message as been added that
   indicates the job was queued again because of a 134 status error.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An imported backup of a backup copy other than copy 1 would base the
   expiration time on the backup time rather than the current time, which
   would result in the imported backup expiring too early.
   

Workaround:
   The bpexpdate command may be used to change the expiration of the imported
   backup.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If the bpschedreq -predict command is executed with a time that is not the
   current day, the exclude dates will not be taken into account.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Generated a much smaller Synthetic Image than a previous TIR incremental
   image when the Synthetic image should have been about the same size as the
   TIR incremental image.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   FlashBackup for Windows does not support backup for cross-mounted volumes.
   This issue has been resolved and backups for cross-mounted volumes are now
   supported.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Several database files could be left open under certain rare error
   conditions.  Over a long enough time, the vmd process could run out of file
   descriptors.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A suspended restore can be restarted more than once if it takes a while
   to transition to the active state.

Workaround:
   Even though the state doesn't change in the job monitor, don't try
   to resume a restore job more than once.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   On servers that do not support xbp, the resource files XNB* were not
   installed.  The effect is not seen on the server, but when pushing client
   binaries to client platforms (using ftp_to_client) that DO support xbp,
   this caused an error because the XNB, XNB.dt were not available for a push.

Workaround:
   After installing on a server, run:
   cp /usr/openv/netbackup/XNB* /usr/openv/netbackup/bin/
   
   Then, push out client bins (such as, rerun ftp_to_client).

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Use of the DISK keyword to specify that an extended copy command be sent
   to a disk involved in the backup does not work with HDLM managed disks
   because the algorithm to convert the disk path to the corresponding sg
   path does not work for the HDLM device.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Description:
   Backup of directories fail.  Changes were made to the compile of bpbkar to
   not use the -dgetdents flag because it does not support the 64-bit file API.

   (NetBackup Clients:  unixware )
--------------------------------------------------------------------------------
Description:
   Collection of disaster recovery information from Windows clients fails.  
   The symptoms of this problem include:
   1) Policy is configured to collect disaster recovery information.
   2) Server connections to the client are supposed to use vnetd.
   3) A firewall is present between the servers and the client.
   When the server attempts to connect to the client to collect the disaster
   recovery information, it will not use vnetd and it will fail.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Auto path correction will down drives that have a valid path, but do not
   respond to commands.  To address this issue, the auto path correction can
   now down drives that do not have valid paths. In addition, this change
   enables ltid to start even with missing paths.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   File promotion is a new feature of VxFS4.0 Storage Checkpoint. Using the
   file promotion feature, a deleted file can be recovered quickly from the
   the most recent checkpoint. Advanced Client Instant Recovery does not
   support VxFS file promotion in the 5.0 GA release.

Additional Notes:
   You can turn off file promotion by touching a file
   /usr/openv/netbackup/DISABLE_VXFS_FILE_PROMOTE.

   (NetBackup Clients:  Solaris Solaris9 )
--------------------------------------------------------------------------------
Description:
   Added the ACS_CSI_HOSTPORT and ACS_SSI_INET_PORT entries to vm.conf.  This
   allows these variables to be set in the vm.conf and used to start acsssi
   when a firewall is between the NetBackup server and the ACSLS server.

   (All UNIX NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Fix memory leak during tar restore of migrated files.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   The restore paths are incorrect for NDMP path-based history.
   
   The list of files for an NDMP backup image are missing the original backup
   path. This occurs with path-based file history running NDMP V4.

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   Enhance bptm to use MEDIA_UNMOUNT_DELAY for duplications to reduce
   unnecessary unmounts/re-mounts during Vault sessions.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Added support for the following Libraries:
   - ADIC Pathlight VX
   - Dell PV-136T and PV-160T
   - HP E-Series Library
   - Qualstar RLS-4221
   - Quantum SDLT600
   - Quantum ValueLdr
   - Quantum ValueLoader
   - Sepaton S2100
   - Sony CSM Library Serialization
   
   In addition, several changes have been added to address serialization,
   library inventory, and extend/retract issues.  The following list
   identifies the library and the issue that has been corrected.
   
   - Dell-PV-136T fixed inventory issue
   - Library Serialization on Sony CSM library series
   - Serialization for HP5597A
   - library inventory issue with Quantum M-Series libraries (missing magazine)
   - extend/retract issue on ADIC Scalar 1000

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The bpvxver script does not always return the proper version of VxVM on
   HP-UX systems.

   (NetBackup Clients:  HP )
--------------------------------------------------------------------------------
Description:
   When automatic path correction is enabled, if the customer has a mix of
   serialized and unserialized robots, some robotic paths may be incorrectly
   changed.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When changing the drive type and the robotic residence of a drive, some
   combinations would yield the error message, "This robot does not support
   multiple media types."

Workaround:
   To avoid this issue, the user can make the required change in two steps:
        1) Change the drive's type.
        2) Change the drive's residence.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   For FlashBackup on Windows, individual file restores of compressed files
   were not restoring the data correctly in all cases. The major problem
   existed when a single compression frame was split between extents,
   something that would occur with fragmented disk allocation.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A 174 error (media manager - system error occurred) occurred when an end-of-
   media status was returned while trying to write a end-of-file marker on the
   tape.  This may occur on Windows or SGI platforms.  The bptm log will show
   an entry similar to the following:
   
   00:30:40.803 [3256.4608] <2> io_ioctl: MTWEOF encountered EOM (error 1100),
   return EOT
   
   This is followed by an entry indicating "buffers out of sequence", and the
   174 error.

   (All NetBackup Servers: Intel-NT Sgi)
--------------------------------------------------------------------------------
Description:
   Two issues caused the NDMP dump levels to always be 1. The issues were:
   
   1 - When merging the STREAMS.alt file into the STREAMS file, the dump
   level of the last backup was over written.  This caused incremental backups
   to repeatedly use dump level 1 instead of incrementing each time.
   
   2 - If there was no previous full backup, the incremental backup dump level
   did not initialize correctly.

Additional Notes:
   This fix was integrated into the 4.5MP path, but the 4.5FP and 5.0 paths
   were missed.  

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The Java restore interface cannot browse files beyond the mount point for
   a point-in-time rollback.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   "I've noticed that ever since using 4.5, you are constantly calling the
   oslevel command, which ends up timing out after so many things are calling
   it at once. For exmaple: > 22340 0 bptm -count -cmd -rt 8 -rn 0 -stunit
   bkmaster05-dlt-robot-tld- > 24564 0 ksh /usr/bin/oslevel > 38992 0
   /usr/bin/rm_mlcache_file > 41868 0 /usr/sbin/instfix -qic -t p What is the
   purpose of this? Can't you call this once when the daemons start up and
   then
   remember it? When this happens I have to find the top level processes that
   died and left an oslevel call hung and kill it off before anything else
   will
   work. "

Additional Notes:
   uname(2) won't work for release level on AIX platforms, (too vague..)
   oslevel(1) needs to be done. Unfortunately, there is (apparently) no
   oslevel(2), and the only way to retrieve this information consistently, is
   to execute the command. ... yuck .. however... cache this value into a
   'static' file indigenous to NB, and use it from there. use
   "my_AIX_ReleaseLevel", and write it to /usr/openv/var (or equivalent) IF
   the file doesn't exist, or is empty, or is corrupt, or is more than 24
   hours old... recreate it. if all else fails, return the sparse string
   provided by the native uname(2)
   
   Further...  the configuration API-s all call oslevel (indirectly) and
   therefore anyone (which implies just about everyone) who accesses bp.conf
   will ultimately invoke oslevel(1), prior to this change.  Thusly this
   change
   will affect many executables.

   (All NetBackup Servers: RS6000)
--------------------------------------------------------------------------------
Description:
   The "bpimagelist -l" command option in the NetBackup 5.0 release
   concatenates the 13th and 14th fields on the FRAG line.  These fields are
   the Tape Allocation Date Timestamp and the Device Written On.  
   
   For example, in the following example the 13th field on the FRAG line is
   11 characters long and is actually the value "1062583209" and the
   value "0" is concatenated.
   
   FRAG 1 1 1856 0 2 12 1 T00000 zrp 65536 0 10625832090 0 *NULL*
   1063188009 0 65536 0 0 0 0 0 0 0

Additional Notes:
   Scripts that use the "bpimagelist -l" output may fail or produce
   erroneous results.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Drives were being downed each time a system-imposed limit on the value of
   a semaphore ID was exceeded causing arvd to exit with an "Error in Semaphore
   Operation" error in /var/adm/message.  

Workaround:
   Down the NDMP drives that are not working.

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   When a synthetic backup is being created, the program bpsynth suspends tape
   volumes that contain images needed for the synthetic backup.  Afterwards,
   bpsynth unsuspends those volumes.  
   
   Bpsynth is suspending and unsuspending even when a volume was originally
   suspended or frozen.  The result is that previously frozen or suspended
   volumes are no longer frozen or suspended after the synthetic backup job
   completes.
   
   To correct this issue, bpsynth does not suspend or suspend tape volumes
   that were originally suspended or frozen or full.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   FlashBackup, either UNIX or Windows, to can now use disk storage units as
   its backup medium.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Description:
   The data image, which is created by synthetic backup, will have a default
   retention level set to two weeks. The retention level that the user has set
   through the configuration was not used.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Synthetic coordinator failed to contact the bpdm process when it finished
   its reading task.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When using synthetic backup on a large number of disk fragments from a disk
   data image, it fails to communicate with the synthetic engine, bpsynth and
   the coordinator, bpcoord.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   FlashBackup for Windows does not support the backup of cross-mounted
   volumes and produces no error message. The backup succeeds but there is no
   catalog information visible for the file system mounted on the junction
   point.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If a client is down when the scheduler is doing autodiscovery, and the
   NetBackup Administration Console for Windows attempts to modify a policy,
   the policy may get deleted.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Performing multiple-map ejects to an ACS robot with more than two LSMs
   could fail if ADJ_LSM entries are not specified in the vm.conf file on the
   media servers connected to the robot.  The default behavior of assuming the
   LSMs are interconnected in a line was not working.  The workaround of
   specifying ADJ_LSM entries is no longer required when the LSM physical
   configuration is interconnected in a line.

Workaround:
   Put ADJ_LSM entries into the vm.conf file on media servers connected to the
   robot describing the physical interconnections of the LSMs.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An NDMP backup restore will fail if you do not use set the mount point
   inode number to 2.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   An Incremental backup, after a synthetic Full Backup, may cause bpdbm to
   core dump.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   A Synthetic Full backup takes too long to create a backup of more than
   20,000,000 files.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Running cron and/or jobs with NetBackup access control enabled requires
   that either the expire date of operating system credentials be lengthened
   using the VxSS utilities, or that accounts be created in the VERITAS
   private domain database with renewable credentials. nbac_cron is a new
   binary that simplifies the process of making unattended scripts NetBackup
   access control aware.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Updated acsssi to include the firewall support from STK for ACSLS.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The tldcd and tl8cd robotic daemons may not be very resilient to having
   connections opened and closed quickly with no data.  This does not affect
   Windows NT systems.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The scheduler cancelled suspended jobs prematurely. The Windows Open File
   Backup does not reuse same snapshot when resuming.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bpjava* executes (UNIX) commands by forking and exec-ing " /bin/sh -c
   '<command>' ". An anachronistic side effect of executing commands, was to
   coerce the form of <command> to be: "LC_ALL=<the locale>; export LC_ALL;
   /bin/ls -l filename" where the intent was to insure that the command was
   executed in the correct locale. This (currently) makes sense ONLY when the
   command is executed from the master server, which is running in its own
   locale and is oblivious to the locale requested by the Java client. The
   user
   server is the ONLY command executed by the master, and the intended locale
   of the user server is passed as an argument, and immediately set after the
   server starts, thus rendering the LC_ALL= environment assignment useless.
   All commands started by the user server will then (naturally) inherit the
   locale, further rendering the LC_ALL= form superfluous. Ordinarily, this
   (old) form is harmless at best, and redundant at worse... except for one
   level of AIX which issues bogus setlocale failures FROM the /bin/sh
   invocation... and spurious output from executed commands is not welcome by
   the Java client... so... remove the locale environment variable definition.
   Historically, it would still be necessary (or an alternate plan devised) if
   the master actually executed commands, more than the user server.  

   (All NetBackup Java GUI capable NetBackup servers and clients)
--------------------------------------------------------------------------------
Description:
   Added capability to run bpbackup with both -i and -L options at the same
   time.  In this case, a progress log will be created on each client machine
   in the policy, as well as on the server running bpbackup.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Added capability to run bpbackup with both -i and -L options at the same
   time.  In this case, a progress log will be created on each client machine
   in the policy, as well as on the server running bpbackup.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When initiating an immediate backup from the NetBackup Administration
   Console for Windows, a warning is displayed if the policy's "go into effect
   time" is too close to the present.  A query for the time on the server has
   been added to allow the user interface to determine the time to see if the
   policy can be initiated or not.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Non-stream in control frozen image jobs were not cancelled correctly when a
   frozen image failure occurred (bpfis error). This resulted in the
   non-stream in control jobs to backup the data multiple times.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Wildcards do not work with restores in multibyte character locales.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NetBackup now uses the following date format when installed on a Japanese
   operating system: 2004/01/19.  This format is consistent with the generally
   accepted format used in Japan.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When doing Oracle proxy copy, the Database (DB) files are treated as sparse
   files.  When the proxy copy backup is combined with local host or alternate
   client off host backup, bpbkar reads the entire file to build the spares
   array. This, for large DB files, can take a long time to finish and cause a
   performance drop in comparison to a standard backup.  

   (NetBackup Clients:  HP Solaris Solaris9 )
--------------------------------------------------------------------------------
Description:
   Oracle Server free with alt-client does not work if multiplexing is
   enabled. This change also improves snapshot error handling during snapshot
   creation and deletion.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If a VxFS file with an extent over 1 terabyte is restored, the restore
   hangs at the beginning and eventually the server times out.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   A cancelled Catalog Backup job leaves the media in the drive instead of
   unloading it and moving it back to its slot. Subsequent use of the drive
   fails because catalog media remains in the drive. Only a manual operation
   from the command line is able to move the media back to its slot and make
   the drive available for further NetBackup operations.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Verifying Network Data Management Protocol drives can fail because the end of
   media (EOM) marker is not recognized. Commands such as bprecover, when used
   with NDMP drives, may fail.

Workaround:
   Use a non-NDMP drive.

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:
   When a checkpointed job resumes, the Activity Monitor displays a 32-bit
   integer instead of a 64-bit integer for the block count.
   
   The bpsched debug log can contain very large values in its ESTIMATE and
   ESTIMATE_STR lines.
   Here is an example:
   log.102703:13:37:09.617 [11222] <2> readstring: ESTIMATE_STR: buf
   (ESTIMATE -1 -1 avocadocat_1067283424) backid 136924 files (-1),  kb
   (18446744073709551615)
   
   This can happen when "Kilobytes Written" is set to its default value, -1.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Restoring from a FlashBackup backup on an HP-UX master server fails.

   (All NetBackup Servers: Solaris)
--------------------------------------------------------------------------------
Description:
   The NetBackup-Java Administration Console fails if NetBackup Access
   Control is used to control access to NetBackup and users who are not
   administrators try to log into a NetBackup server on a Windows system.
   Users who are not administrators do not have the appropriate permissions
   to write to the log directories and files on the Windows system.

Workaround:
   Non-administrator users must have write permission to the NetBackup log
   directories and files on the Windows system. The workaround is documented
   in the NetBackup 5.0 Release Notes for UNIX and Windows, "Operational Notes,"
   "NetBackup Java Interfaces," "General Notes," item 5, Windows 2000 NetBackup
   host access from NetBackup-Java interface as a non-administrator.

   (All NetBackup Java GUI capable NetBackup servers and clients)
--------------------------------------------------------------------------------
Description:
   The NetBackup jobs database daemon (bpjobd) may fail when the bpdbjobs
   command, with the -report and -all_columns options, is used to query the
   database.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When resuming a suspended backup job with Windows Open File Backup enabled,
   the resumed backup will not re-use the original backup job's volume
   snapshots. Instead, if the WOFB Error Control configuration setting was set
   to Abort on Error, the resumed backup job will abort and a status 156-
   snapshot error will be encountered. If the WOFB Error Control configuration
   setting was set to Disable Snapshots & Continue, the resumed backup job
   will be resumed without any volume snapshots. This behavior is not correct
   and the resumed backup job should re-use the original, suspended backup
   job's snapshot volumes if they were successfully created.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Occasionally, on RedHat Advanced Server 3.0, NetBackup does not detect
   drives going from ready to not ready and then to ready again with a
   different tape label.

   (All NetBackup Servers: Linux)
--------------------------------------------------------------------------------
Description:
   A number of manual steps are required in order to prepare hosts to use
   NetBackup Access Control and in configurations involving a large number of
   hosts, and a means of automating some of the preparation would ease the
   initial configuration process.
   
   The script "nbac_install_helper" installed in the goodies directory
   provides basic support for automating the initial Access Control
   configuration steps listed in the "Access Management" section of the
   System Administrator's Guide. An executable version is also provided for
   installations unable to take advantage of the scripted version.
   
   Documentation describing use of the script is provided by "-usage" and "-
   help"
   flags, by comments in the source code of the script, and by run-time notices
   when configuration is attempted.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When resuming a suspended backup job with Windows Open File Backup enabled,
   the resumed backup will not re-use the original backup job's volume
   snapshots. Instead, if the WOFB Error Control configuration setting was set
   to Abort on Error, the resumed backup job will abort and a status 156-
   snapshot error will be encountered. If the WOFB Error Control configuration
   setting was set to Disable Snapshots & Continue, the resumed backup job
   will be resumed without any volume snapshots. This behavior is not correct
   and the resumed backup job should re-use the original, suspended backup
   job's snapshot volumes if they were successfully created.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When resuming a suspended backup job with Windows Open File Backup enabled,
   the resumed backup will not re-use the original backup job's volume
   snapshots. Instead, if the WOFB Error Control configuration setting was set
   to Abort on Error, the resumed backup job will abort and a status 156-
   snapshot error will be encountered. If the WOFB Error Control configuration
   setting was set to Disable Snapshots & Continue, the resumed backup job
   will be resumed without any volume snapshots. This behavior is not correct
   and the resumed backup job should re-use the original, suspended backup
   job's snapshot volumes if they were successfully created.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When using the Java GUI to initiate drive cleaning, the drive names are
   truncated after 22 characters.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The NetBackup scheduler may core dump when it prepares to exit if it is
   trying to determine which storage units are busy.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Corrected numerous situations where a backup job could not be
   restarted/resumed because the necessary information was not retained. The
   most visible situations were the storage unit not available (219) errors,
   but there were a number of other conditions.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Suppose the master server and media server are different machines.  Suppose
   that the media server and client are the same machine so that shared memory
   backups and restores are normally done.  Suppose the master server is
   configured so that "connect backs" with the media server are done via
   VNETD.  The typical bp.conf entry on the master server would look
   something like this:
   
   CONNECT_OPTIONS = mediax 0 1 2
   
   where "mediax" is the hostname of the media server.  The digit "1"
   signifiies that vnetd style
   connect backs are used.
   
   In this case, restores will fail.  Typically, the job will fail with error
   code 5 and be marked as
   "incomplete" in the job monitor:

Workaround:
   Don't use vnetd style connect backs with the media server if the media
   server is a client.
   

   All NetBackup Windows Clients All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Customers using noauth authentication with the windows GUI would not be able
   to
   start manual backup jobs if they were not an Administrator on the system
   running the
   windows GUI.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   VMD on any volume database host can corrupt the volDB (media database), and
   may core dump.  This is due to an error with the pool database, introduced
   with NetBackup 5.0.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Restores of Windows paths containing "\" don't work.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Bpdbm failure in SQL agent browse for FI backups

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Scheduler started multiple jobs instead of just one when restarting a job.
   This situation occurs when multiple multi-copy jobs are submitted for
   restart/resume. The GUI sends all the selected job IDs, but scheduler needs
   to perform a many to one mapping of the requests. Filtering is added to
   scheduler to perform this mapping.

Workaround:
   select only one job of the multi-copy job set when performing a
   restart/resume

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   VxFS file promotion does not support alternate location restore. A check is
   required for Instant Recovery restore, for alternate location restore, file
   promotion method needs to be turned off.
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Add recognition for the ACS_LTO_200G vm.conf entry.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   email notification of backup status is not being sent for Windows clients.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Multi-stripe SQL RFI backup isn't displayed in SQL image browser. It
   appears that bpbrm provides XBSA a bad restore list, setting the
   xbsa_flag to XBSA_DATA_EXTENSION for some objects that do contain data.
   The case occurs on multi-stripe retrievals, where the first set of
   objects contain data, the next set contains no data and the third set
   do contain data. After the second set starts using the
   XBSA_DATA_EXTENSION (-3), it continues to be used for the third set
   which do contain data.
   
   This test case relies on a backup done by washer to
   CAM: /washer.MSSQL7.WASHER.db.Ye Olde Howard's Bar and
   Grill.~.7.001of003.20040224135654..C
   

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Some conditions will cause bpplschedrep to crash when setting last day of
   the month.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   If a multi-stream Windows backup is suspended or cancelled prior to
   going "active" (for example: while the snapshot is being taken by stream
   number one), then it's possible the suspended/cancelled backup can not be
   resumed/restarted.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Jobs failing with a 134 status were not requeued.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Performing backups using an exclude list with non-fully-qualified path
   names could cause significantly slower backup performance.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   tldd/tl8d dumping core when attempting to kill an orphaned child process.  
   Only happens when running NetBackup Media Manager in VERBOSE mode.

Workaround:
   Remove the "VERBOSE" entry from your vm.conf file and do not start ltid
   using the -v option.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   cancel job vxvm snapshot backup hanging

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Jobs ending with 134 and 59 exit codes were marked as incomplete and
   cancelled. correct operation would be to retry the job.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Multi-stream & multi-copy jobs fail if all storage units are of type disk

Workaround:
   Use at least one tape when performing multi-copy backups

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   The vfm.conf and vfm_master.conf files do not list FlashBackup as a
   valid policy type as of 5.0. This was supported at 4.5FP3 and our
   documentation (and developers) state that FlashBackup is supported with
   VVR at 5.0. However, when 5.0 was shipped FlashBackup was omitted from
   the VVR fim section.

   (All NetBackup Servers: HP Solaris)
--------------------------------------------------------------------------------
Description:
   libVnbconf and libVnbatST reported "Text file busy" when the
   client binaries were being laid down on an AIX machine.  This
   is likely to "someone" making a bpcd request to the
   media server while the install was occurring.  bpinst had been changed
   in 5.0GA to avoid this issue but it was not expected that we needed
   to change the other client pushing scripts too.  Updated cp_to_client,
   ftp_to_client, install_client, client_config and move_libs to run
   slibclean, to put the affected files in *.new and then move them to
   their real locations.
   

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   PFI SQL images are not expiring properly

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   When the customer submits multiple jobs to a Disk STU with Max Jobs set
   to 1 all the Submitted jobs go active.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   
   On HP 11x PARISC systems, some 32bit executables were reporting
   unresolved symbols when libVxSS_helper.sl was trying to be dlopened.
   
   shl_findsym(/usr/openv/lib/libVxSS_helper.sl,_shlInit)/usr/lib/dld.sl:
   Unresolved symbol: typeid__XTQ2_3std9exception_ (data) from
   /usr/openv/lib/libVxSS_helper.sl/usr/lib/dld.sl: Unresolved symbol:
   __nwa__FUlRCQ2_3std9nothrow_t (code) from
   /usr/openv/lib/libVxSS_helper.sl/usr/lib/dld.sl: Unresolved symbol:
   __nw__FUlRCQ2_3std9nothrow_t (code) from
   /usr/openv/lib/libVxSS_helper.sl/usr/lib/dld.sl: Unresolved symbol:
   __dt__Q2_3std9exceptionFv (code) from
   
   The creation of this shared library was changed to link with two
   C++ libraries.

Workaround:
   If you are using database agents on a 32-bit HP/UX machine, rename the file
   /usr/openv/lib/libVxSS_helper.sl.  For example:
   
   cd /usr/openv/lib
   mv libVxSS_helper.sl libVxSS_helper.sl-5.0GA
   
   

   (NetBackup Clients:  HP )
--------------------------------------------------------------------------------
Description:
   Unable to cancel, resume, or suspend  using bpdbjobs

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Resuming a checkpointed backup job resulted in selection of a disk storage
   unit instead of a tape storage unit. This situation can arise at a site
   that has a media server with both tape and disk storage units. A resume
   needs to continue on the proper media. If initially tape, then resume on
   tape. If initially disk, the resume on disk.

Workaround:
   Don't mix disk storage units and tape storage units on the same media
   server.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Cannot resume Windows multi-streamed backup jobs after error 219 (storage
   unit not available).

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   Running backups using TIR with Move Detection can leave a hung bpbkar
   process when any errors occur during the job.

   All NetBackup UNIX Clients
--------------------------------------------------------------------------------
Description:
   Synthetic Backup image is bad and will not restore.
   When extent ends as fragments ends, synthetic engine didn't handle it
   well. Additional data is copied and causing catalog mismatch with image.

   (All NetBackup Servers)
--------------------------------------------------------------------------------
Description:
   NDMP backup fails writing TIR data.
       
   bptm log shows:
   19:56:36.854 [2969] <16> write_data_tir: ERR - Cannot write TIR data to
   media, NDMP return code 9.
       
   ndmp logs shows:
   
   19:56:36.801 [02969] 162988 ->  (8) 267 [0] 19:56:36 NDMP_TAPE_WRITE
   (0x304) NDMP_NO_ERR (0x0)
   19:56:36.801 [02969] 162988 .   data_out(0)=
   19:56:36.801 [02969] 162988 **  End NDMP_TAPE_WRITE
   19:56:36.850 [02969] 162988 <-  (8) 599 [267] 19:57:51 NDMP_TAPE_WRITE
   (0x304) NDMP_NO_ERR (0x0)
   19:56:36.850 [02969] 162988 .   error=NDMP_ILLEGAL_ARGS_ERR (0x9)
   19:56:36.850 [02969] 162988 .   count=0 (0x0)
   19:56:36.850 [02969] 162988 **  End NDMP_TAPE_WRITE
   
================================================================================